Category	Type	Preview	Tags	Description
fantasy	15-to8highkey		effects	TO8' High Key LORA (SD1.5) by theovercomer8 Link to original page:https://civitai.com/models/9205?modelVersionId=11464 Lora trained bytheovercomer8 using the technique for diffusion with noise offset. Used to produce beautiful high-key images in stable diffusion. Varying the model weights can affect the blown-out lighting effect
anime	22h		semi-real .base models comics fantasy illustrations environments	A great model for beginners, 22h aka "Estilo Vintedois" (meaning Style 22) is the product of Brazilian model creators Predogl and piEsposito. Use it to create fantasy characters and illustrations without a lot of complicated prompting. It is similar to the Dreamlike base model, with an oil paint and CGI aesthetic. Brilliant for fantasy animals, video game assets, as well as 2D illustrations. The included token is estilovintedois. {VERSION:V0.2}
effect	2dsprite		gaming effects fantasy	2D Sprite style by Creativehotia 2D Sprite Style is a concept artwork model for videogames, by model creator CreativeHotia. As the name implies, the art looks flat like those early high-quality iPad videogames with isometric view by default, which you can modify by pairing it with a base model like <revani>. The recommended weight is 0.6 and there are four triggers to modify the effect: Assets, Character, Armor, and Weapons. The creator recommends working in 786x786 with inpaint and control net, as well as combining it with other LoRAs for more varied effects and asset creation.
sci-fi	3d-cg-realistic	No		No
sci-fi	7-of-9		characters female sci-fi fantasy realistic star trek	7of9 by Hbait2 The borg female character from Star Trek played by Jeni Ryan. Works well with realistic base models like <believe>
style	80s-style		art style retro	kEighties by konyconi This lora introduces an artistic style based on the 80s.
effect	90s-photo		effects lighting retro	90s flash photo by coyote
Animal	a-chan		female characters anime	No
anime	aba-guilty-gear		characters female gaming monster girls anime	No
anime	abubu-style		anime hentai art style	Abubu Style Lora by RedChocobo This lora introduces an artistic style based on the artist Abubu. Use weights of ~0.8.
pose	accidental-exposure	No		No
anime	acrosstable-pov		poses environments anime	pov across table concept by emberbyte
clothing	adjusting-clothes			Adjusting clothes by psoft
effect	adv-enhancer		effects detailers lighting realistic	Advanced Enhancer LoRa by Z_phyr Truly a Swiss army knife. It removes unnecessary detail, adds the necessary one, works three times harder with the texture of skin, eyes and hair, if a portrait is generated. Correct overlay of light and shadow, reflections and glares on surfaces. Perspective correction. The algorithm of pyramid noise offset, working in both directions at the same time, allows to achieve a tonal range as close as possible to real photographs. LoRA knows many names of photo equipment and types of lighting, so feel free to use professional terms of photographers and videographers in prompts. It is recommended to use the weight up to 2 inclusive. If you think that the level of detail or noise offset is still insufficient, then use it together with other LoRA that solve these problems ( because increasing the weight more than 2 can lead to unexpected results ).
anime	adventure-time-princesses		anime hentai cartoons characters female monster girls	Adventure Time Princess Pack by ricketysplit C'mon, grab your friends and set your weight below 1 and a base like <aohard> or <av45> to use it. There are 9 subjects here which you'll need to call by the trigger names. Moderl creator RicketySplit says that "it' pretty reliable at telling them apart, but you may need to explicitly mention features like skin color or accessories sometimes to push it in the right direction, especially if you mention more than one character."
fantasy	adventurers		fantasy gaming clothes	Adventurers by Hatch Adventurers is a sexy-themed fantasy clothing LoRA that boosts detailed clothing effects in traditional RPG classes. Use [negative prompts] to control how much boobage and youth it puts out. Also suitable for boosting the details in videogame characters. The recommended weight is 0.65. Important: It may glitch if the character LoRA is set very high, so prepare to lower both weights to find the sweet spot
character	aespa-karina		korean asian characters female realistic	AESPA Karina by alelele
clothing	after-shower		clothes porn	After Shower by EDG
pose	ahegao		expressions anime porn	TongueDrop (ahegao): a LoRA in the ShirtLift universe by brandnewfakeuser
anime	ahegao2		expressions anime porn	Ultimate Ahegao by MoistDD
anime	ahegao3		expressions anime porn	Ahegao iuui style by alphano
anime	ahegao4		expressions anime porn	Ahegao by Satyam_SSJ10
anime	ahegao5		expressions anime porn	Ahegao by ALKHEM Recomended on upper body, portrait or head focus poses Works pretty well at weights ~0.8
inversion	aid210		inversion .negatives .negatives	Useful Quality Embeddings - AnimeIllustDiffusion by Euge are a Huge help in boosting image quality. Example: /render portrait of a lovely woman [[<aid210:-1.5>]] <revani> Adjust the weight between -0 and -2. Make sure it is always put in the negative prompt or very bad things happen
base	aio		.base models realistic semi-real fantasy	Project AIO as in All-In-One is a blended checkpoint from multiple realistic and semi-realistic fantasy models, focused on character portraits, sci-fi and overall diversity. It' good at a lot of things, except hands, says model creator Ndimensional, who also says to try it with panties pulldown LoRAs...
lora	akatsuki-outfit	No		No
celebrity	alicia-schmidt		female realistic characters	Alica Schmidt by jadimaster A realistic likeness of model Alicia Schmidt. Use with a realistic base like <believe> or <realvis20> at a weight of 0.7 or similar
object	alien-spaceships		sci-fi environments objects spaceships	C.A.G. - Alien Spaceships v1.0 by LDWorksDavid The LORA has two activations, one without token and with it. # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low # Mangled limbs, general chaos ? Some LoRas can’t handle rendering with resolutions bigger than 512x512 // Also make sure your prompt isn’t too complicated for this LoRa # No visible effect ? Check if the LoRa has any triggers that need to be added in order to activate it # Good full models to try out LoRas with: anime ? <anylora> <counterfeit> <breakanime> comics ? <comics-vision> <icomix> <misstoon> semi-real ? <revani> <lyriel16> <dreamshaper6> realistic ? <majic> <bra5> <chilloutmix> <edge>
pose	allfours-pov		poses hentai fetish	?Pov?All Fours and From Above v1.0 by rerorerorero # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low # Mangled limbs, general chaos ? Some LoRas can’t handle rendering with resolutions bigger than 512x512 // Also make sure your prompt isn’t too complicated for this LoRa # No visible effect ? Check if the LoRa has any triggers that need to be added in order to activate it # Good full models to try out LoRas with: anime ? <anylora> <counterfeit> <breakanime> comics ? <comics-vision> <icomix> <misstoon> semi-real ? <revani> <lyriel16> <dreamshaper6> realistic ? <majic> <bra5> <chilloutmix> <edge>
effect	amateur-porn-style	No		No
lora	ame-chan-ngo	No		No
lora	ana-de-armas		characters female realistic latin	Ana De Armas?LoRa? by dogu_cat Ana Celia de Armas Caso is a Cuban and Spanish actress who starred in the popular drama El Internado. Try the lora with real base models like <realvis20> between 0.7 and 1
celebrity	anadearmas		characters female realistic latin	Ana De Armas?LoRa? by dogu_cat Ana Celia de Armas Caso is a Cuban and Spanish actress who starred in the popular drama El Internado. Try the lora with real base models like <realvis20> between 0.7 and 1
pose	anal-dildo		porn poses	Anal Dildo LORA by Quiron a lora that creates renders of anal dildo insertion Recommended settings for getting good renders: # Guidance at approx 7.0+ # Best weight around 0.1 - 0.3 # Works well with realistic models and non-realistic models # Works well with basic and elaborate prompts Sidenotes: # Seems to interact well with other loras # Render at low sizes and use /more /size /strength to upscale # This lora can be a little finicky, you may only get one usable render per batch Example Prompt: /render (Dutch angle), (AnalDildoQuiron style),1girl, blonde woman, anal toy, dildo, anal, (anal object insertion), <anal-dildo:0.1>[blurry, lowres, worst quality] <realflex>
pose	anal-pov		poses porn	POV Doggy Anal + Creampie LoRA by KinkAI A lora that creates renders of anal doggy position and creampie. Put anime and amateur into negative prompts. Use from behind prompt for reverse cowgirl position. Recommended Lora strength 0.7 - 0.8.
pose	anal-povrv		poses porn butts	POV Reverse Cowgirl Anal + Creampie LoRA by KinkAI A lora that creates POV reverse cowgirl anal images. Recommended Lora strength 0.7.
pose	anal2-pov		poses butts porn	Close Up Anal + Creampie LoRA by KinkAI A lora that creates close up shots of anal sex in various poses and angles. Recommended Lora strength 0.7 - 0.8.
style	analog		realistic .base models realistic art style	An older Stable Diffusion 1.5-based reality model, it was one of the first high quality realism photographic dreambooth models and gained notoriety for it' touch of retro. Works well on portraits of people and animals, has a hazy film quality like a manual Hollywood camera. It hasn't aged well as it is a 512x512 model, but it' still aesthetically cool
style	analog-madness5	No		No
celebrity	andrew-tate		male characters white	Andrew Tate by MyNameIsThis
effect	angel-punk		effects art style	AngelicAI by konyconi This lora is designed to add angelic effects to absolutely anything. A sportscar or combine harvester will suddenly sprout wings and a halo, a toaster will gain an ethereal aura. Works great with revani and at default weight ????
fantasy	angelic-warriors-valkyrie		anime characters female armor	Angelic Wairrors (Valkyrie, Paladin, Priestess) by Hatch This lora is designed to generate Valkyries, Paladins, and Priestesses, with secondary options for Knights and Angels. Use cleavage or prompts about breasts to get breast armor.
celebrity	angelina-jolie		characters female realistic white	Angelina Jolie?LoRa? by dogu_cat It' the actress, prime years (95-06). This LoRA feels overtrained, sometimes scrunches. Use it with a base concept and not as a standalone lora, like <angelinajolie:0.7><realvis13> or <deliberate2>
background	anime-backdrop		objects effects interiors anime	Role Background PRO by 1175114098100
background	anime-bedsheet		anime	Dakimakura design / Bed sheet / Pillow by Tomas_Aguilar This is a dakimakura style wide print design for an anime bed sheet, a more wide version of the other model we have called <anime-pillowcase>. The recommended dimenions for this one is 512x1080. Intended for nsfw waifus, but it also generates people relaxing in bed unless you're prompting something spicy with a base like <aohard>
anime	anime-eyes		anatomy expressions anime	Eye - LoRa by Kokoboy Glittery, dramatic anime eyes effect. Try it at a weight of 0.8 and go down from there if it is overkill. Can be used in extreme closeups to draw fantasy shapes inside the eye, add tears, and so on. The base model <revani> is frequently seen in examples along with <av45> and <aom3> and <meinap> For the "jewel eyes" effect, setup your prompt with words like: masterpiece, best quality, depth of field, 1girl, eye focus, close-up, crying, green hair, jewel, star eyes, green, forest, etc.
effect	anime-festivalcard		anime effects	Love Live! School Idol Festival Cards by Ars_Arcanum This LORA aims to replicate the style and layout of Love Live! School Idol Festival Cards Best resolution is 512x720 Use <av3> as the base, and a lora weight of 1 or lower Additionally you should use one of the rarities together with their detail level: N rarity, simple background R rarity, simple background SR rarity, simple background SSR rarity, detailed background UR rarity, detailed background And lastly use one of the color attributes: pink attribute blue attribute green attribute Using the name of one of the main cast' girls also replicates them quite well. Some issues: Smaller CFG Scale creates fancier outfits while bigger CFG creates simpler designs It seems to struggle with hands especially with smaller CFG values The color of the attribute often seems to bleed into eye, hair and clothes colors The small symbol on the bottom right often only shows up after Hires. fixing The names are never readable
effect	anime-graphic-design		anime graphic design art style	Cover Page Layout Graphic design by yakineko This is intended for modern anime layouts of a single character with abstract poster-like elements and stylish (but totally illegible) kanji characters. The base model examples shown are <meinap> and <majic5> so try with any range of semi-real and 2d anime models at a weight between 0.5-0.7. Increasing the weight appears to add larger text
anime	anime-jc		effects anime hentai art style expressions	JCM2 - Style by JohnJohn A whimsical, snarky anime style inspired by cartoonist JCM aka JCM-2. The signature look is a youthful, snarky woman with a mischievous facial expression. Thus, try these in your prompt for best results: freckles, smirk, naughty face, serious face, (color) eyes, looking at viewer. You can also choose to focus the background or not, by selecting [blurry background, soft focus] as your negative. A good base is <cetus>, <aom3> or <aohard> Note: This LoRA is unpublished by the model maker.
anime	anime-minimalist		anime art style effects	Minimalist Anime Style by hortacreator Anime Minimalist creates simple 2d images that almost look like paper cutouts, in interesting pastel and contrasting colors. Works great with a weight of 1 and a base like <av5>. You may want to lower the weight when stacking it with other LoRAs
effect	anime-nostalgia		anime effects lighting	[LoRa] Nostalgia/ ?? Concept (With multires noise version) by L_A_X Nostalgia is a LoRA that used to be called "OldTime" which is designed to do two things: tint the colors to warm pastels to give a feeling of an era long past, and to also add "light and delicate backgrounds" which the latter appears to be largely predicated on the base model you choose. Try it out with base models like Ghost Mix or similar. The model creator L_A_X recommends a lora weight of 0.8
anime	anime-pillowcase		anime	dakimakura (pillowcase) by uhustir Dakimakura Pillowcase is an anime styled tall / vertical template to generate your waifus on. It works best with dimensions like 272x832, and pair it with your favorite anime base.
object	anime-pillowsex		objects anime hentai porn	Masturbation / pillow humping / female focus / dakimakura by Tomas Agilar Masturbation / pillow humping / female focus / dakimakura is a Lora that works well with <aohard> at default weight or lower.
style	anime-realism		art style detailers	Anime<>Realism Control by advokat This lora can strengthen the 2D style or photo style depending on the strength. Negative strength to strengthen 2D style (up to -1.0) and positive values will strengthen the photo style. Works well with blended models. Range of values depends on which models, generally stick to range of -1 to 1, and scale accordingly
effect	anime-realism2		detailers	BreakRealize LoRA v.rl2170 by br d A lora for making your picture more or less realistic. Using it with minus weight like <break-realize:-1> will lead to more illustrated outcome, while doing the opposite will provide more 2.5D looking renders. For stronger effect use high weights like: <break-realize:1.5> <break-realize:-1.5>
style	anime-shobo		art style anime hentai trans male penis porn realistic lgbtq	Aya Shobon Style - LoRA by TroubleDarkness An anime style intended to emulate the artwork of artist Aya Shobon. Main features: Monstrous futanari penises Somehow normal futanari penises Saggy / Perky breasts Twitching Testicles (TBD) Recommendations: 1.0 - 0.8 weight for most models Relevant genital tag tips: You can adjust tag weights by placing it in parenthesis like ((Bulge)) So, Large Erect Penis and ((Large Erect Penis)) would produce different size of schlongs. The second one also can be much bigger or equal to Huge Penis. Things that works pretty well: Erection \| (Size) Erect Penis \| Flaccid \| (Size) Flaccid Penis for penis state (Size) value could be Small \| Large \| Huge or Gigantic Large \| Huge \| Gigantic Penis for penis size Large \| Huge \| Gigantic Testicles for testicles size Sagging \| Resting Testicles for testicles state Sagging for saggy testicles Resting for on-surface state while sitting (like breast rest) Futanari + Bulge for bulges (nothing more needed) Female Pubic Hair \| Excessive Pubic Hair \| for pubes Things that works somehow: (Size) Semi-Erect Penis \| (Size) Floppy Penis for penis state Actually the same tag, the second one not to match danbooru tags Full-Package Futanari \| Futanari + Pussy for penis and pussy Might be useful for From Behind view Backsack for testicles from behind Might be useful for From Behind view. Important: LoRA won't produce futas all the time (just don't prompt futa/penis), so you could generate regular clothed or naked girls
effect	anime-tarot		effects anime art style	Anime Tarot Card Art Style LoRA (???/???????) by CyberAIchemist Gives a tarot-card like background effect to your image. Tested with anime-based models. Use the following trigger words: (details about your background) background, tarot card to change the background and style used. Generations will not be the same as civitai due to missing embeddings/textual inversions, so may need some fine tuning. Strength of LoRA recommended from 0.4 - 0.8, but adjust if need to.
style	anime30s		comics cartoons art style retro	LoRa 1930' Cartoon RichStyle 1.5 by RICHVIP
style	anime80s		anime art style retro	Akemi Takada (1980s) Style LoRA by Lykon Akemi Takada is a Japanese manga artist and anime character designer born in Tokyo, Japan. Her character-design work is one of the more recognizable styles of the 1980s, when she worked on series like Creamy Mami and Kimagure Orange Road. This LoRA model is useful for recreating the likeness of heroine characters like Ayukawa Madoka. Use the triggers to modify the art style, like watercolors
style	anime90s		anime art style retro	1990s Anime Style LoRA by Lykon An ink and marker classic style based on popular 1990s anime, however note that it requires a strong negatives to look like the sample pictures. Works with most popular anime base models. The recommend weight is around 0.5-0.7
style	animelineart		anime art style illustrations art style	Anime Lineart / Manga-like (??/??/????/???) Style by CyberAIchemist Anime Line Art is more along the lines of traditional Manga than just line art. Setting a lower /guidance will make the lines thinner, says creator CyberAlchemist. Try pairing it with an anime base model like <aom3> or <av45> at a weight of 1 or higher
effect	animepop		anime lighting effects	Contrast Pop Art (LORA) [FFXIV RESHADE PRESETS] by duskfallcrew Animepop is the best way to describe DuskfallCrew' experimental anime contrast filter, which is trained on Neneko GSHADE — a popular Final Fantasy mod tool that added contrast to the game. Try it with 0.6 weight and an anime base model like <av5>. It is limited to portraits and will freak out into abstract shapes with certain base models
pose	animesplit		anime hentai poses	Split by halfawave An anime-style standing split position, intended for base models like <aom3> and <aohard>. The creator Halfawave notes that it doesn't work well with realistic models (will result in random open leg positions instead), and some users noted that it will make women' legs more thick and/or athletic regardless of your prompt
pose	animesplit2		anime poses	[Concept]High legs by Eisthol
anime	anna-frozen	No		No
celebrity	anya-taylor-joy		characters female white	Anya Taylor Joy by Holex_P
style	anylora		anime .base models semi-real	Let' get one thing out of the way:AnyLora is an anime base model that doesn't work with every lora. The "Any" part refers to the Anything series of anime base models, and that' what it' most commonly used for. Created by Lykon, the same model creator that makes Dreamshaper, AnyLora focuses on high compatibility with many anime LoRAs in a safe range of aesthetics. Works well with other models by Lykon, like <neoartcore>
inpainting	anylora-inpainting	No		No
style	aoaoko-pvc		.base models anime toys	This unique base model Aoaoko (also its creator' name) is useful for creating anime figures. There are also toy and other anime figure LoRAs in our system, so try a combination both! Here are the notes from the creator: This model aims to generate high quality anime girls with PVC figure texture. It is strongly recommended to add [worst quality, low quality:2] to the negative prompt for a more realistic effect.
celebrity	aoc		characters realistic female latin	Alexandria Ocasio-Cortez (AOC) by MachineMinded Alexandria Ocasio-Cortez, also known by her initials AOC, is an American politician and activist. She has served as the U.S. representative for New York' 14th congressional district since 2019, as a member of the Democratic Party
style	aohard		anime .base models hentai	AbyssOrangeMix2 - Hardcore by Havoc This is the more NSFW version of <aom3> but uses Orange Mix 2 as its base instead. This has the better VAE for more contrast. Can also do semi-realistic oil paint style like AOM3. This update removed the 'black box' issue some users were experiencing.
inpainting	aohard-inpainting	No		No
base	aom3		anime .base models illustrations	Abyss Orange Mix 3 - High quality flexible model - abyss orange mix 3 (A3) with kl-sf-8 vae. "a midpoint of artistic but bends towards anime - the model has been tuned to combine realistic textures, a artistic style that also feels like an oil colour style, and a cute anime-style face. Check out our detailed guide for prompts and tips: https://www.animemaker.ai/abyss-orange-mix-3-guide/
celebrity	aoyama-hikaru	No		No
anime	aponia		characters anime gaming semi-real honkai	Aponia ( Honkai Impact ) by Trauter LoRA trained on 3 different clothing styles of Aponia, our mommy, one of the flamechasers of the Previous Era in Honkai Impact 3. Good for anime bases and semi-real bases. Strength is from 0.1 - 0,7. To apply the 3 different styles, use the following keywords: For Aponia Battlesuit, use tags: aponia_outfit, 1girl, blue eyes, breasts, large breasts, looking at viewer, nun, veil, black dress, breast curtains, detached sleeves, long sleeves, frills, pelvic curtain, sideless outfit, thigh strap To add her signature butterfly wings, add this tag: butterfly dress For bikini: aponia_bikini, aponia, 1girl, breasts, large breasts, white headwear, earrings, hair ornament, butterfly ornament, frilled swimsuit, frills, one-piece swimsuit, criss-cross halter For casual outfit: 1girl, blue eyes, breasts, large breasts, casual outfit, office wear For OG Outfit: aponia, aponiaoutfit, 1girl, breasts, large breasts, blue eyes, mole under eye, eyes visible through hair, mole, nun, blonde hair, long hair, bangs, hair between eyes, veil, black dress, breast curtains, detached sleeves, long sleeves, frills, pelvic curtain, sideless outfit, butterfly dress Now go, and render for all that is beautiful in the world.
character	arab-doll		characters female white asian	Doll Likeness by EDG Maghreb (NorthWestern part of Africa) aka Arab Doll Likeness is a LoRA portrait model of beautiful West Arab women, generally including Morocco, Algeria, Tunisia, and such areas. It' an olive-skinned woman with dark hair. Use it with a weight of 0.7 or less, with a realistic or semi-real base model like <dreamshaper6> or <lyriel>
anime	arca		characters .base models art style	This is a custom model we trained based on just a few images of @tlack305' pomeranian, named Arca. She' a haughty bitch at times but we love her very much.
style	arcane		art style semi-real gaming fantasy	Arcane Style LoRA by Lykon Not to be confused with the older Arcane model, this is a new LoRA that can be combined with other base models. The model creator Lykon says he did not tag specific characters from the show, its simply an art style inspired by Arcane
celebrity	ariel-winter		characters female realistic	No
celebrity	arielle-maxine		lgbtq female characters trans	No
anime	arknights-eureka		anime female characters	Arknights-U-Official (Eureka)-Last All-Arounder Beauty Girl by L_A_X
anime	arknights-surtr		characters anime female	????-???? Arknights-Surtr by zbw Character lora focused on Surtr, one of the operators from the game Arknights. It’s easy to use and works well with most models. Using it without any triggers will make the character gain Surtr’s characteristics, while each trigger is responsible for one of her outfits. Avoid using multiple triggers simultaneously, as this may confuse the AI. It will try to split one person into two/three, each with different clothes. Any weight works - regulate accordingly to your own preferences.
anime	arknights-texas		characters anime female asian	Arknights-Texas the Omertosa by L_A_X Character lora focused on Texas, one of operators from the game Arknights. It’s easy to use and works well with most models. It can also be safely combined with a lot of other loras. Using it without any triggers will make the character gain Texas’ characteristics, while each of the triggers introduces small changes to the pose or background. Any weight works - regulate accordingly to your own preferences.
inversion	arm-grab-doggy-ti			No
inversion	armor		armor clothes fantasy	Armored Clothes by DevilKKW. Use it in a positive prompt like this: /render /tall professional photograph of woman wearing tiger (<armor>) in an outdoor setting. sharp detail, detailed skin, sneer, pores, wrinkles, cinematic, soft lighting, high quality, film grain, Fujifilm XT3, shallow depth of field, natural light [[<verybad-negative:-2>]] <photon>
clothing	armorsuit		fantasy clothes armor	Armor Suit(????) LoRa by axebro Mostly flexible Lora, creates loads of different kinds of armor. Sweet spot for PD is around 0.1-0.5 weight. Get crazy with it and you can make some unique armor. A lot of the really cool prompts have other loras combined with this one.
pose	armsupport-pov		anime poses	?POV?Arm Support and From Below ???? by rerorerorero
anime	aroma-sensei		characters female anime	No
style	art-deco-punk		effects art style architecture interiors	ArtDecoAI by konyconi This lora is designed to produce images based on the art deco artstyle and design movement. It has a strong effect and little prompting is needed to achieve the desired effect. It will try to add the style to even the most mundane items. Works fantasic with revani and default weight.
style	artgerm		art style comics illustrations oral. porn oral anime	Artgerm (Stanley Lau) style Lora by Lykon Should work fine at around weight 1 on AnyLoRA ckpt. and also super well on Animes. Use artgerm to increase fidelity. To make his marker style use traditional art. For full body pics remember to use a Detailer or face inpaint. Remember this is a style LoRA. To make characters accurately you need other networks too (but it can get pretty close with accurate descriptions).
style	arthemy		.base models anime cartoons comics illustrations art style	Arthemy Comics v4 by Aledelpho Arthemy Comics (previously known as Comics Vision) is a full checkpoint base model, a blend of models like MissToon and Icomix. As such, it serves as a good western-style illustration base model that' right on the edge of anime. The optional trigger Toontanium can alter the results, but is not included by default. For best results, render directly at 512x768 and then /facelift /anime with the ones you like. Here' prompting advice directly from the model' creator, Aledelpho: Positive: (artwork:1.2), (lineart:1.33), CHARACTER \| SETTING \| EMOTIONS, (hyperdefined), (inked-art), COLORS \| AESTHETIC, complex lighting, (flat colors), ultradetailed, (fine-details:1.2), absurdres, (ATMOSPHERE) Replace BIG LETTERS with adequate prompts of your own choice. Negative: lowres, blurry, (worst quality:1.3), (sketch), (text,words:1.3), watermark, (simple background), glitch, (jpeg-artifact:1.2), desaturated, compressed
pose	asian-vaginas			No
lora	ass		butts porn anatomy fetish	JK Big Ass Slim Waist (PAWG ver.) by jk_chstr
pose	ass-from-wall			Ass On Wall (Concept) by HolmesCage A reverse glory hole style ass coming out of a wall, with the character' picture frame above it. Truly specific
pose	ass-on-glass		butts porn poses	Ass On Glass LoRA by schneed
pose	assgrab-pov		butts porn poses	POV Ass Grab by ekune
sci-fi	atom-punk		sci-fi effects fantasy	Atompunk (Rethinking) by natinitakira
style	atoz		semi-real .base models fantasy anime environments art style	A-Zovya RPG Artist Tools v3.0 by Zovya On one hand, A-to-Z RPG can do quite a bit as a general purpose full base model for fantasy lovers. That said, quite a number of PirateDiffusion community members have pointed out that the examples shared by the creator are unattainable without a lot of trickery. As such, we want to warn beginners not to start here, or consider getting a rough first draft with <atoz> and either remixing into another model or heavily applying LoRAs for a polished look. Recipe #fantasy can also give you quick results. UPDATE v3 Compared to v2, the new version of Atoz is one step closer to fantasy than anime. The author also states that it contains more RPG knowledge which should make creating fantastical creatures easier.
celebrity	aubrey-plaza		characters realistic female	Aubrey Plaza by td-discord507 A likeness of comedian and actress Aubrey Plaza. Try it with a base like <believe> or similar realistic base model
inversion	autumn		environments	Instant autumn by devilkkw. Put it in your positive prompt with a weight of 1 or higher, its easy to use
style	av3		anime .base models illustrations	Many cite Anything V3 one of the first universally loved anime models, and with good reason. This self-described "latent diffusion model for weebs" is one of the most popular classic 2D anime models, for its crisp looks and easy prompting. It has been succeeded and blended into various other models, including the controversial Anything 4 series by a different team. This one is the more cell-shaded aesthetic, while the later versions have more semi-real shading. The origins of the model is a fascinating rabbit hole...
style	av45		anime .base models illustrations	This is not the original Anything V4.5, which many of our users felt the color was "way too gray" due to the VAE (or lack thereof) on the seminal repo. Most liked this one way more: a repack called AV45 Colorful Pro by model maker Manseo, which produces much brighter images, and excels in that semi-real anime style
style	av5		anime .base models illustrations	Anything V5 is the harder to use, better for Loras successor to AV3. For most people, you'll want to use AV3 and get nicer pictures. If you're going for accuracy, AV5 might be worth your time. Though iterative but not revolutionary update, it' certainly pretty. Mix it with Loras using guidance 7
celebrity	awkwafina		characters female asian realistic	Awkwafina by pyroflare77
celebrity	ayase-haruka		asian japanese characters female	AyaseHaruka_JP_Actress by pmeantweetanthony
character	azami		female characters semi-real	Azami2 v1 by fanskllll Recommended settings for getting good renders: # Weight ~0.15-0.3 # Try it out with base models such as <chilloutmix> or <majic> # /parser:new /clipskip:2 (recommended)
anime	azurlane-bremerton		anime characters azur-lane female boobs	Bremerton All Outfit 5in1 (Azur Lane) (???? ???? ???) by RiuKi_MK1 An explicit LoRA model of the anime character Bremerton from Azur Lane. Check the triggers for different outfit options. Works best at low weights, if your image looks fuzzy try a different base model too. Realistic is also possible
anime	azurlane-formidable		anime characters female azur-lane	formidable?Azur Lane? by JHU48 Formidable, also known as HMS Formidable, is the name of a supporting character of an aircraft carrier (aka ship waifu) in Chinese strategy game Azur Lane, a side-scrolling shoot-em-up simulator RPG. It also spawned a 12 episode anime series and is released on PS4 and Switch. Her signature looks are selectable as triggers, though it looks like her without them. Try it with <perfectworld> (semi-real) base or an anime base like <av3> or <meinamix>
anime	azurlane-gali		characters female anime	La Galissonniere (Azur Lane) Cat v2.0 by rerorerorero General lora guidelines: # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low # Mangled limbs, general chaos ? Some LoRas can’t handle rendering with resolutions bigger than 512x512 // Also make sure your prompt isn’t too complicated for this LoRa # No visible effect ? Check if the LoRa has any triggers that need to be added in order to activate it # Good full models to try out LoRas with: anime ? <anylora> <counterfeit> <breakanime> comics ? <comics-vision> <icomix> <misstoon> semi-real ? <revani> <lyriel16> <dreamshaper6> realistic ? <majic> <bra5> <chilloutmix> <edge>
anime	azurlane-st-louis		azur-lane characters anime hentai female	St. Louis (Luxurious Wheels) (Azur Lane) by Bloodsuga
anime	azurlane-taihou		azur-lane characters female anime hentai	Taihou (All Skins) (Azur Lane) by Bloodsuga
style	babes		semi-real anime .base models cartoons illustrations interiors architecture	Creates hourglass-figure pinup cartoon female characters, like the proportions of Jessica Rabbit. Can be bent to create CGI as well. For a more cartoony look, try the icomix + babes mix called <comicbabes>
style	babes2	No		No
effect	back-light		environments effects	?? Enhanced Backlighting by AlchemistW Enhanced Backlighting is a Lora used to improve the backlighting and sidelighting effects in environments with strong lighting contrast. Many models have not adjusted their noise_offset, which can result in an overly balanced overall brightness and a lack of emphasis on dark areas or photographic reflection effects, especially for 2D models. Even when the prompt for backlighting is given, the character still appears to have the same brightness as the environment, lacking a prominent highlight and making the image appear very "flat" in tone. This LORA intentionally deepens the contrast to improve this problem and has a natural light illumination improvement effect for both 2D and 3D models. Try it at a weight of 0.5 with your favorite base model as a safe starting point. Your /guidance setting may also affect the effect, so adjust both accordingly
lora	background-details		detailers backgrounds	Background Detail Enhanced or Simplified by ritansonoac554 can significantly affect the adherence of the background. When the LoRA is used with a negative weight, the background will appear softer and blurry. When the weight is high, the background becomes sharply focused and more details can be seen. It' a good compliment to the other anime-styled lora <distance-control> Try it with a realistic base model like <majic6> or your favorite anime base.
background	background-tweak		detailers	Background Complexity Additive by Jemnite This LoRA helps to increase the complexity and prominence of your backgrounds. This reduces the style, coloring, clothing, anatomy or facial details. Play around with positive and negative weights.
object	backwards-cap	No		No
sci-fi	bacterium		sci-fi creatures effects	djz Bacterium by driftjohnson
inversion	bad-artist	No		Bad Artist negative inversion by Nick The Hacker. Put this inside a negative prompt to boost the quality of the image. Example: [[<bad-artist:-2>]]
inversion	bad-artist-anime		.negatives	No
inversion	bad-chilloutmix		.negatives	Bad Picture 75 Vector Version for Chilloutmix by SeekerOfTheThicc is a negative embedding designed for the <chilloutmix> base model. Use it like this: a portrait of a woman [[<bad-chilloutmix:-1.5>]]<chilloutmix> To adjust the effect, choose a weight between -0 and -2
inversion	bad-dream	No		Bad Dream by Lykon is a negative embedding to boost quality. It must be placed in the negative prompt with a negative weight. Example: [[<bad-dream:-1>]]
inversion	bad-hands		.negatives	No
inversion	bad-hands-v5		.negatives	No
inversion	bad-pictures		.negatives	Bad Pictures Textual Inversion by Vivipeg is a versatile negative embedding to help you produce higher quality images. It works better when there are less than 75 tokens in your prompt. Example: a portrait of a woman [[<bad-pictures:-1.5>]] <dreamshaper8> To adjust the effect, use a weight between -0 and -2
inversion	bad-prompt	No		Bad Prompt negative embedding by Nerfgun3 is a general detail booster, which can fix details like hands. To use it, put it in your negative prompt with a negative weight like: [[<bad-prompt:-1>]]
sci-fi	badass-cars		vehicles fantasy sci-fi	Badass Cars by Barmagloth
celebrity	baesuzy		characters realistic female	Bae Suzy by alelele A female character resembling South Korean singer Bae Suzy. It was trained on an older model photo realism base model called VisionGen, which is most similar to <aio> and <clarity>
anime	bags-under-eyes		anatomy expressions anime	concept Bags under eyes(dark circles) by Saya
lora	bald		hair memes	Hairstyles Collection by antonio_riolo2610
clothing	balinese-kebaya		clothes	Balinese Kebaya by MeepMeepow
object	ball-gag		objects bondage oral	Ball_Gag by frief A Lora that produces photos of characters with a ball-gag in their mouths Recommended settings for getting good renders: # Keep guidance low! About 4-8 # Weight 0.1 is the sweet spot # Works with most realistic based models (best with <deliberate2>) also works with un-realistic models Sidenotes: # Seems to interact well with other loras including character loras # A good resolution to render at is /size:640x640
inversion	ball-gag		bondage fetishes porn	No
lora	ball-gag2		bondage fetishes porn	No
clothing	ball-gown		clothes	Haute Couture \| Ball Gown by EDG
pose	balls-kick		memes fetish anime poses	Ball Kick ( Femdom ) by Torque
object	banknote		effects memes objects anime	money style ?? by kozue Puts your character on paper currency. Seems designed for anime and humor. Creator recommends a resolution of 960x448 and a weight of 0.5-0.8 with <av45> or similar
sci-fi	barbarella		fantasy sci-fi characters female retro white	Barbarella by rogueAI
lora	batgirl		female characters comics realistic semi-real white	No
lora	batgirl-killing-joke	No		No
clothing	battle-suits		sci-fi clothes mecha anime fantasy objects armor	Parrley' Battle Suits by EDG
clothing	bavarian-dirndl		clothes	Bavarian Dirndl by EDG
anime	bayonetta		characters semi-real female asian	Cereza Bayonetta by Creativehotia Cereza Bayonetta is a character similar to the video game. Try it with a weight of 0.7
style	bb95		furry fantasy semi-real .base models	BB95 Furry Mix v9.0 by Bumbleblee95 One of the best furry models out there. It' a mix of various models and can generate various anthro ideas. It' doing well in generating photorealistic males and females. SFW and NSFW. Example positive: anthro (white wolf), male, adult, muscular, veiny muscles, shorts, tail, (realistic fur, detailed fur texture:1.2), detailed background, outside background, photorealistic, hyperrealistic, ultradetailed, Seems to work better with /clipskip:1 If you know names of good furry artists, throw them in to improve results.
clothing	bdsm-hood		objects clothes porn bondage	hood by l19850831 A black hood with a mouth cutout for bondage roleplay. Looks good with realistic models like <realvis20>
pose	bdsm-leash		poses bondage	BDSM - On a Leash LoRA by KinkAI a lora that creates renders of people on a leash Recommended settings for getting good renders: # Guidance at approx 7.0+ # Best weights from 0.7 to 1 # Works great with realistic models and non-realistic models # Works great with basic and elaborate prompts Sidenotes: # Seems to interact well with other loras # Render at low sizes and use /more /size /strength to upscale Example Prompt: /render 1girl, solo, beautiful redhead, leashed <bdsm-leash:0.8>[blurry, lowres, worst quality]<edge>
pose	bdsm-outfit		bondage poses fetish	a lora that creates renders of women wearing bondage gear Recommended settings for getting good renders: # Guidance at approx 7.0+ # Best weight right around 0.65-1.0 # Works great with realistic and non-realistic models # Works great with basic and elaborate prompts Sidenotes: # Seems to interact well with other loras # Render at low sizes and use /more /size /strength to upscale Example Prompt: /render /sampler:k_euler_a /size:512x768 nsfw, 1girl, solo, 20 year old blonde woman, bondage outfit, <bdsm-outfit>, [bra, underwear, blurry, lowres, worst quality] <vivy>
pose	bdsm-rope		objects poses porn bondage	Bondage Suspension by Gtonero A general bdsm pose and object modifier that works on real, semi-real and anime images. You can also prompt for what the arms are doing (bound, arms up, etc). However, the creator warns not to use long prompts (many tokens) and to increase the strength of this when other LoRAs are used. For example, Try: <bdsm-rope:1.4><kdoll:0.2><chilloutmix>
pose	bdsm-rope2		poses bondage fetish	a lora that creates renders of women in Japanese shibari rope bondage Recommended settings for getting good renders: # Guidance at approx 7.0+ # Best weight right around 0.65-1.0 # Works great with realistic and non-realistic models # Works great with basic and elaborate prompts Sidenotes: # Seems to interact well with other loras # Render at low sizes and use /more /size /strength to upscale Example Prompt: /render 1girl, solo, face, nsfw, nude, rope, shibari, arm up, pussy, <bdsm-rope2> [blurry, lowres, worst quality] <vivy>
pose	bdsm-tied-bed	No		No
pose	behind-pov		porn poses butts	POV from behind by hoshi119
base	believe		.base models realistic fantasy	"I can't believe it' not photography", final version, by ResidentChiefNZ. This is a mega-merge model that brings in 10 different realistic base models into one. The creator suggests setting guidance between 4-7, and recommends it for CGI characters and landscapes. Vae included is SD-v2 840000
inpainting	believe-inpainting	No		No
lora	belle-delphine		porn characters female	Belle Delphine by DiffusedIdentity
lora	belle-delphine-ultimate	No		No
pose	below-pov		porn poses	From Below [POV] by Satyam_SSJ10
pose	bentovertable		poses porn	Bent over table + Creampie LoRA by KinkAI Tag suggestions: leg up, leg down, from side, from behind, flat, close up, vaginal, anal, creampie, cum. Include things like 'bending over table' or 'getting fucked by man from behind' in your prompt. Recommend a minimum resolution of 640x640, any lower and you will get bad faces. Recommended LoRA strength 0.7-0.8, based on URPM
celebrity	berta-castane		characters female realistic	A likeness of 20 year old Spanish actress Berta Castane, created by Bozack3000
effect	betterbodies		anatomy realistic	breastInClass: Better Bodies by throwawayyy111 The model name is Breast In Class: Better Bodies. You get more idealistic NUDE female body types out of generated images. It' best at low weights with accompanying keywords. 0.3 to 0.5 is good for most purposes. Very high weights can cause the face to crop out. Very high weights can cause the image to burn. Adding "small breast" or "huge breast" keywords can add more size variation, keeping the shape. Adding stubborn loras or conflicting terms (e.g. "shirt") to the prompt can burn the image. It alters the body shape, particularly the breasts. The data set is a collection of adult women with their faces cropped out of the photo. What it is not: The dataset is all naked women with no faces. Any effect on clothing, men, flat chests, or faces is unintended. For a base try <chilloutmix> or similar. Works with semi-real models as well
clothing	bikini		clothes boobs	JK Tight Bikini Top [Underboob + Skindentation] by jk_chstr
clothing	bikini-cowprint		clothes anime	Cow Print and Bikini by Goofy_Ai
clothing	bikini-micro		clothes	JK Eyepatch Bikini Top by jk_chstr
clothing	bikini-micro2		clothes	Eyepatch bikini by Gtonero
clothing	bikini-micro3		clothes	Miniature Microbikini by wewewew
clothing	bikini-mini		clothes anime semi-real realistic clothes anime realistic semi-real	Ex micro bikini by buranj Introduces a shorter micro-bikini for your renders. Token: micro-bikini. Useful for anime, semi-real and realistic models. Renders using realistic models may not be so good since this was designed for anime and semi-real in mind
clothing	bikini-slingshot		boobs clothes anatomy	bikini by wop1359602 Introduces a slingshot micro-bikini for your renders.
celebrity	billie-eilish			Billie Eilish Pirate Baird O'Connell (born December 18, 2001) is an American singer and songwriter. She has a history of political activism, focusing on climate change awareness and women' equality. Do NOT use the trigger words 'Billie Eilish' in the prompt since the result may be influenced by possible base model training data. The lora was trained with the trigger word 'nomobil' instead. Prompt example: nomobil with blonde shoulder length hair and bangs with blue eyes and brown eyebrows Explore the lora with a starting weight between 0.6-0.8
lora	bimbo-3u	No		No
object	bipedalmech		mecha armor objects	Bipedal Tactical Mecha by stormriver Use "mecha28"as a trigger in your prompts. Most of the bipedal mecha LoRAs here were of the fantasy genre, appearing excessively pristine for warfare weaponry. Therefore, I trained a LoRA model based on majicMIX realistic v6, hoping to make the generated mechas more rugged. I used weight 1.0 for this LoRA.
character	bismuth		semi-real .base models realistic realistic	BismuthMix is a merge of Chillout, Deliberate, DDos, El Zipang, and Retmix. It' an underrated gorgeous model right on the edge of reality, but not quite: It looks like high-end Unreal Engine renders. Use it for colorful portraits, particularly daytime and interior shots. Produces great bokeh. To boost quality, use positive prompts like (moody portrait, striking features, beauty, intricate details, dramatic composition, tension, wispy hair, blue eyes, contrast, texture, realism, high-quality rendering, stunning art, high quality, film grain, Fujifilm XT3, detailed skin) and add flaws like acne, blemishes, and freckles to make your characters more believable
lora	biting-lips		anatomy expressions	{DD} Biting lips SFW semirealistic/realistic version by aDDont Introduces a concept of lip biting for your renders.
pose	bj-ass-co-op		oral porn poses butts	concept Cooperative fellatio anilingus by Saya
pose	bj-bang-pov		poses porn oral penis	concept Blowbang Lora by SysDeep Point of view blowjob where many dicks are pointed at the woman on the floor. A strength of 0.5 - 0.8 with a URPM base works best
pose	bj-coop-analingus		penis oral porn butts	No
pose	bj-deep		poses hentai porn porn oral penis	The key here is things like balls_deep, deepthroat, irrumatio, hands on head, and so on. I focused on images with the balls_deep tag
pose	bj-deep-frames		poses hentai oral penis porn	concept deepthroat before during after by emberbyte
pose	bj-deepthroat		porn poses oral	MS Real Lite - Deepthroat LoRA by Murky a lora that creates deepthroat bj renders Recommended settings for getting good renders: # Guidance at approx 7.0+ # Best weight right around 0.8 - 1.0 # Works great with realistic models and non-realistic models # Works great with basic and elaborate prompts Sidenotes: # Seems to interact well with other loras # Render at low sizes and use /more /size /strength to upscale # Yes, I know, ANOTHER BJ lora, but this one does what it does very well and very simply Example Prompt: /render 1girl, looking at viewer, 1boy, hetero, nude, penis, solo focus, pov, oral, fellatio, hand on another' head, pov crotch, deepthroat, <bj-deepthroat> [blurry, lowres, worst quality] <believe>
pose	bj-double		poses realistic oral penis porn	DoubleDick by FKGOD A porn model for a woman on the floor holding two dicks. Check the optional triggers to trigger the different effects, and use <urpm13> or newer for the base
pose	bj-frames		poses art style oral penis porn	{DD} Deepthroat frames for gif by aDDont A point-of-view generator of deep blowjob images, creating templates (non-animated) that can be turned into GIFs with another program. Check the triggers for options. Lora weight of 0.75 and a base like <deliberate> should work well
pose	bj-implied		poses hentai porn	(Concept) - Implied-Fellatio/Implied Fellatio by LazyTrainer Implied fellatio lora, from behind. Try it with the <aohard> base, though seems to do well on semi-real models as well. Set to 0.6-0.9 strength
pose	bj-pierce-pov		oral porn poses	Licking my dick piercing tongue by dalixyz This lora creates images of women with tongue piercings giving oral sex. it works fine with very simple prompts but the weight can really be no higher than <bj-pierce-pov:0.02> or it begins to break down. It works with both realistic and non-realistic models. Adding prompts like "from above", "from side" and "pov" will help refine your final result.
pose	bj-pov		poses oral penis porn	(POV) Blowjob + Creampie LoRA by KinkAI This LoRA is for creating "POV" and "side view" blowjob images with cum-in-mouth functionality as well as deepthroats. Recommended weight is 0.8. Some triggers only work in either POV or side view. For example, 'Arms behind back' only works in side view. Based on UberRealisticPornMerge, so start with <urpm13> or <degen>
pose	bj-reverse			https://civitai.com/models/18543/concept-reverse-fellatio concept Reverse fellatio by Saya
pose	bj-side		oral porn poses	Fellatio (Side View) by ekune This is a LoRA to generate Fellatio (Side View). Recommended weight range from 0.7 - 1.0
pose	bj-side-frames		oral porn poses butts	{DD} Blowjob from the side frames for gif by aDDont
pose	bj-sloppy		hentai poses porn oral penis	Sloppy Fellatio LoRA by schneed This LoRa is prone to blowjobs viewed from the side rather than a POV of the crotch. If you want a crotch POV, put 'pov, crotch pov' in your tags. The weight is good from 0.6 to 0.8. Sometimes, the images produce extra fingers/fingers without arms on penises. For this, inpainting will work to remove them. If the base model has a problem with penises, hands, and fellatio adjacent subjects, it will have problems with this.
pose	bj-stomach		oral porn poses	{DD} Fellatio on stomach, legs up by aDDont "a lora that creates renders of women giving oral sex while lying on their stomach Recommended settings for getting good renders: # Guidance at approx 7.0+ # Best weight right around 1 # Works great with realistic models and non-realistic models # Works great with basic and elaborate prompts # This is an excellent lora that performs great right out of the box, VERY robust Sidenotes: # Render at sizes similar to landscape or 768x512 to allow the lora to work properly # Seems to interact well with other loras including character loras # Render at low sizes and use /more /size /strength to upscale # When using this with anime based models decrease the lora weight or the renders will tend to be more realistic (try starting around 0.4) Example Prompt: /render /size:768x512 <bj-stomach:0.75>, fellatio, blowjob, on stomach, 1boy, huge penis, (1girl:perfect face, cute, small breasts, long ginger hair, petite) [blurry, lowres, worst quality] <edge>"
pose	bj-toh		porn poses oral	???? Blowjob Top of Head POV Lora???? by Quiron Yes, ANOTHER oral sex lora. This time she just takes the head in her mouth. Recommended settings for getting good renders: # Guidance at approx 7.0+ # Best weight between 0.5 and 1.0 # Works great with realistic and non-realistic models # works well with basic and elaborate prompts Sidenotes: # Seems to interact well with other loras # Render at low sizes and use /more /size /strength to upscale # Multiple penis can be solved by using "penis"as a negative prompt Example Prompt: /render BlowjobTopofHeadPOVQuiron pose, 1girl, oral,penis, <bj-toh:0.6> [blurry, worst quality, lowres] <paragon>
pose	bj-triple-pov		poses porn oral penis	POV Triple Licking Oral by ekune This is a LoRA to generate POV Triple Licking Oral. The creator suggests using this at 0.8-1.0 weight for best results but make sure to always play around with the weight. It was generated and only tested using AOM2, AOM3 A1, OrangeMix VAE and AnythingV3 VAE. Also suggest using 512x512 then upscaling the picture to an higher resolution. If you add triplets to the prompt - You can possibly generate triplets which is hype. This Lora was made on a very small dataset of 15 images, so don't expect a ton of versatility... but when it works, it works
pose	bj-worship		oral porn poses	Cock Worship / Alternative Deepthroat Model 1.4 by grendar
pose	bj2-pov		poses porn oral penis	Redo of POVBJer by honk_honk V3 UPDATE: This update greatly expands the promptable camera angles and adds or improves several other concepts. Note the 22 triggers. Here are some notes from the creator, Honk Honk: Poses: squatting, kneeling, all fours, lying on stomach (no lying on back yet) Camera angles: from front, from above, side view (stricter), from side (gives a wider range of side angles) Penises of Color: brown cock, dark brown cock (you sometimes need to add e.g. dark-skinned male or black man to avoid a default-tone body, and titrate strength until it makes him the desired shade without effecting his blower) Sideways blowjob, cock in cheek, cheek bulge Contains bit of 2boys or 2girls training, recommend kissing a cock or licking a cock instead of sucking a cock for multiple women, no consistent solution yet for e.g. one on shaft and one on balls. Works great on most models and both converts anime to realistic and realistic to anime. Of course, works great with no character added as well. Effective at nice low weights of 0.2 - 0.4 with a base like URPM Some comments on the model noted difficulty using with character Loras, so your mileage may vary — and if you do, set the character weights very low, too
anime	black-cat-marvel	No		No
anime	black-widow-marvelwhatif	No		No
character	blackdoll		characters female black realistic semi-real anatomy	Pretty Black Girl by Ciuron This lora helps create good dark skin tone without distortion, trained on the faces of cute black women. It was trained on the base model <aio> (all-in-one) and works well at a weight of 0.8. It also works well with some semi-real and anime models
pose	blacked		porn hentai	BLACKED by psoft No trigger words, standard guidance, regular weight. Trained on NAI, should work with most models. You don't need LoRA specific trigger words, just as long as you have include male tags in your prompt (e.g. 1boy, 2boys, multiple boys) and/or sex-related tags (e.g. sex, fellatio, gangbang). In some models you can increase the effect by using tags such as black man, interracial, etc. The idea was to achieve good results with the least impact on style and composition as possible. You can use this LoRA in combination with other LoRAs (as long as those are not overtrained). Depending on the model, author suggests between 0.8-1.1 weight and standard guidance.
clothing	blacked-underwear		clothes	Blacked underwear - clothing by psoft This is a simple outfit LoRA. Author tries to make it as low impact as possible on the character, style and composition. Can use weight above 1.0. If you use the word 'blacked' it helps generate more accurate text on the underwear. It can be used with various types of clothing but the basics are: sports bra and thong.
celebrity	blackpink-jennie		korean asian characters realistic female	JENNIE For BLACKPINK by 378866459393
effect	blackred		lighting effects	Bloodstained - Vector / illustrative by machinedelusions Black & Red is a LoRA model called Bloodstained, which we renamed as to not cause confusion with actual blood stained horror LoRAs. This is less of an art style and more of a lighting effect, it will push illustrations, semi-real, and real photos into a black and red color scheme. Bring on the weight, over 1 is fine
inversion	blade-runner		art styles sci-fi fantasy	Blade Runner textual inversion by PirateKitty. Usage: Add <blade-runner> to your positive prompt with a base model like <dreamshaper8>
anime	bluearchive-asuna		characters female anime blue archive	Ichinose Asuna (??????) Blue Archive \| Character Lora 1630 by Numeratic Ichinose Asuna v1.0 by Numeratic # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low # Mangled limbs, general chaos ? Some LoRas can’t handle rendering with resolutions bigger than 512x512 // Also make sure your prompt isn’t too complicated for this LoRa # No visible effect ? Check if the LoRa has any triggers that need to be added in order to activate it # Good full models to try out LoRas with: anime ? <anylora> <counterfeit> <breakanime> comics ? <comics-vision> <icomix> <misstoon> semi-real ? <revani> <lyriel16> <dreamshaper6> realistic ? <majic> <bra5> <chilloutmix> <edge>
anime	bluearchive-blacksuit		characters male anime blue archive	Black Suit/??/??/?? ??(blue archive) by HCSJ Black Suit v1.0 by HCSJ # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low # Mangled limbs, general chaos ? Some LoRas can’t handle rendering with resolutions bigger than 512x512 // Also make sure your prompt isn’t too complicated for this LoRa # No visible effect ? Check if the LoRa has any triggers that need to be added in order to activate it # Good full models to try out LoRas with: anime ? <anylora> <counterfeit> <breakanime> comics ? <comics-vision> <icomix> <misstoon> semi-real ? <revani> <lyriel16> <dreamshaper6> realistic ? <majic> <bra5> <chilloutmix> <edge>
anime	bluearchive-isami		characters female anime blue archive	Isami Kaede?????????? Blue Archive by Kyakusaki Isami Kaede v1.0 by Kyakusaki # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low # Mangled limbs, general chaos ? Some LoRas can’t handle rendering with resolutions bigger than 512x512 // Also make sure your prompt isn’t too complicated for this LoRa # No visible effect ? Check if the LoRa has any triggers that need to be added in order to activate it # Good full models to try out LoRas with: anime ? <anylora> <counterfeit> <breakanime> comics ? <comics-vision> <icomix> <misstoon> semi-real ? <revani> <lyriel16> <dreamshaper6> realistic ? <majic> <bra5> <chilloutmix> <edge>
anime	bluearchive-koharu		characters female anime blue archive	CatEyes Koharu (Blue Archive) v.lion by pip_install_civitai This LoRA only shows a specific expression of Koharu! Most eyes also fail! Be patient and keep generating images. General lora guidelines: # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low # Mangled limbs, general chaos ? Some LoRas can’t handle rendering with resolutions bigger than 512x512 // Also make sure your prompt isn’t too complicated for this LoRa # No visible effect ? Check if the LoRa has any triggers that need to be added in order to activate it # Good full models to try out LoRas with: anime ? <anylora> <counterfeit> <breakanime> comics ? <comics-vision> <icomix> <misstoon> semi-real ? <revani> <lyriel16> <dreamshaper6> realistic ? <majic> <bra5> <chilloutmix> <edge>
anime	bluearchive-rin		characters female anime blue archive	Rin (Blue Archive) v1.0 by Goofy_Ai # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low # Mangled limbs, general chaos ? Some LoRas can’t handle rendering with resolutions bigger than 512x512 // Also make sure your prompt isn’t too complicated for this LoRa # No visible effect ? Check if the LoRa has any triggers that need to be added in order to activate it # Good full models to try out LoRas with: anime ? <anylora> <counterfeit> <breakanime> comics ? <comics-vision> <icomix> <misstoon> semi-real ? <revani> <lyriel16> <dreamshaper6> realistic ? <majic> <bra5> <chilloutmix> <edge>
inversion	blueprints		illustrations	PlanIt! Blueprints by Mousewrites supports additional detailed labels and schematic notes and multiple angles of an object. Try it a base model like <revani>
effect	blur-background		effects	Tangbohu BackGround Blur LoRA (???????LoRA) by TangBohu This is a LoRA used to control background blur while maintaining the original image as much as possible; Increasing the weight will gradually blur the background, while reducing the weight will make the background clearer, with a maximum/minimum of 3/-3! More upper limits can be explored. When using larger weights, there may be significant changes in the image
effect	blur-foreground		effects	Tangbohu foreground ??? Blur LoRA (??????????LoRA) by TangBohu This is a LoRA used to control foreground blur while maintaining the original image as much as possible; Reducing the weight will gradually blur the foreground, with a maximum/minimum of -5! More upper limits can be explored. When using larger weights, there may be significant changes in the image
anime	bmo		characters cartoons creatures	BMO (Adventure Time) by quack_quack_22 BMO is the green Gameboy-looking character from the Adventure Time cartoon. Getting facial details right will depend on which base model you choose and how close-up BMO is. Any anime or comics base that is 1.5-SD based should work well, like <aohard>
clothing	body-con-dress		clothes	Bodycon Dress >Customizable Colors< ??° (LoRA) by Elluna The Body Con Dress is a woman' garment that has long sleeves and soft material like thin cotton sweater, but is cut as a tight form-revealing one piece very short dress with an open v-neck to accentuate the chest. It' quite different to the Virgin Killer, as the material is more thin and isn't open on the sides or sleeves. Use this with a very low weight between 0.3-0.5 and a base like <chilloutmix>. Some users noted that it will change the face at higher weight ranges
effect	body-horror		horror effects creatures	Body Horror Creatures by Sibience This lora generates creatures similar to the one from ‘the thing’ movie. There aren’t any trigger words but it reacts well to the prompt "disgusting body horror creature". You can change the word "creature" for something else like "person" or "dog", but in some models it will always try to add human forms, most likely because of the word "body" in the prompt. At high weight levels it can create some gross stuff but at lower ones and combined with other loras it’s capable of adding surprisingly nice creature effects. It was tested on a few different models and seemed to work well for both realistic and anime stuff.
clothing	body-stocking		objects clothes fetish hentai	????????\|seamlessbodystocking by anystone
clothing	body-stocking2		clothes fetish porn	bodystocking by ZoeHeart
lora	bondage		fetish bondage porn	BDSM by hoshi119
lora	bondage-and-gag		poses hentai objects bondage fetishes porn	Bondage and gag by halfawave Suggests AbyssOrangeMix2 with Anything V3. Author used between 0.3 and 0.7.
effect	bones		effects horror	BoneyardAI by konyconi The Boneyard AI Lora adds bones to everything. It works so well with the default settings, we were left speechless. Try it with realistic or illustration base models and use it on people, persons, and places. Make bone coffee machines, bone shoes, bone fetish porn, bone anime hats, whatever. Bones! A weight of 1 is fine, and a flexible base like <revani> may do better with your more weird ideas
object	bongs		objects	Bongs by Eggbena This lora, created by a community member, was designed to create bongs with a realism style. It was found to work best with realistic base models. Lora was trained on product photos, people smoking as well as a few used bongs. In hindsight, many of the bongs used for training were not filled with water. Prompting in water bong or water pipe can help with this slightly.
lora	bored-sex-gaming		memes gaming porn	Disinterested Prone Sex Gaming \| Sex Act Lora 66 by Numeratic A staple theme in some hentai mangas - a bored girl playing on a handheld console while getting banged from behind. This LoRa works best with anime-based models but is hard to use. Especially fingers, it may take some patience and time to get them right. Use weight 0.7-1 and experiment with sizes like 640x640, 704x512, etc. The author suggested using <aohard> to start working with it. Normally does nintendo switch, may be able to do controller if prompted correctly. Example prompts that may help: blue hair, yellow eyes, holding, holding handheld game console, nintendo switch, bottomless, indoors, on bed, simple background, emotionless sex, distracted,
inversion	boris-vallejo		art styles paintings fantasy sci-fi	No
lora	boruto-eida		naruto anime characters female	No
lora	boruto-hanabi		characters female anime naruto	No
lora	boruto-hinata		characters female anime naruto	No
lora	boruto-sakura		characters female anime naruto	No
celebrity	boruto-sarada		naruto characters female	Boruto Sarada by Pleased_Chomusuke
lora	boruto-temari-nara		naruto anime characters female	No
pose	boundandgagged		poses hentai objects bondage	Bondage and gag by halfawave Suggests AbyssOrangeMix2 with Anything V3. Author used between 0.3 and 0.7.
anime	bowsette		anime characters gaming super mario monster girls	Bowsette \| Character Lora 1860 by Numeratic LoRA trained on Bowsette from super mario. Use tokens:bowsette to trigger. Works for anime and photo-realistic bases. Strength recommend from 0.4 - 1, but feel free to experiment
style	bra6		.base models asian japanese korean chinese realistic female	Better Realistic Asians version 6 by BankaiPlease is the last release before the SDXL-based model. This update increases the number of faces possible though some noted that it may be more challenging to prompt for outside of the creator' templates. Check the info link for those prompt templates
style	breakanime		.base models anime architecture illustrations	BREAK Anime is a very easy to use anime model that generates great results with minimal prompting. It' also a good choice for mechas, buildings, and detailed drawings. Compared to <breakdomain>, this model is better at everything that' even slightly connected to anime - especially girls. The colors are also more vivid and saturated.
style	breakdomain		.base models anime architecture illustrations	BREAK Domain is a model focused on very detailed backgrounds. This is the latest in a series of models that are trying to introduce a new prompting divider 'BREAK' so that tokens before and after this break are handled differently, if it actually works... Example "absurdres, highres, ultra detailed, (1girl), BREAK , kaleidoscopic imagery, symmetrical patterns, vibrant colors, geometric shapes, mesmerizing designs, optical illusions, dynamic composition BREAK , pantomime art, expressive body language, silent storytelling, evocative gestures, visual narratives, theatrical performances BREAK , blooming flowers, colorful petals, fragrant scents, nature' bounty, vibrant gardens, peaceful scenery" Compared to <breakanime>, this model is better at semi-realistic renders, especially fantasy themed ones. The colors are also more muted and not as saturated.
style	breakreal		.base models anime architecture illustrations	BREAK Domain is a model focused on very detailed backgrounds. This is the latest in a series of models that are trying to introduce a new prompting divider 'BREAK' so that tokens before and after this break are handled differently, if it actually works... Example "absurdres, highres, ultra detailed, (1girl), BREAK , kaleidoscopic imagery, symmetrical patterns, vibrant colors, geometric shapes, mesmerizing designs, optical illusions, dynamic composition BREAK , pantomime art, expressive body language, silent storytelling, evocative gestures, visual narratives, theatrical performances BREAK , blooming flowers, colorful petals, fragrant scents, nature' bounty, vibrant gardens, peaceful scenery" <breakreal> is the last one of the breakdomain series. This one is, as the name suggests, oriented towards a more realistic style. With appropriate prompts it can get really close to life-like illustrations.
lora	breast-squeeze		female porn poses boobs	This lora creates images of women squeezing their breasts together Recommended settings for getting good renders: # Guidance at approx 7.0+ # Best weight around 1.0 # Works well non-realistic and realistic models # works well with basic and elaborate prompts Sidenotes: # Works well with other loras # Render at low sizes and use /more /size /strength to upscale Example Prompt: /render 1girl, solo, ((arms down, elbow squeeze, breasts squeezed together)), <breast-squeeze>, [lowres, blurry, worst quality] <photon>
effect	breasts		anatomy porn boobs	NSFW Natural Breasts by SettingsNoProblem is a large breast modifier to add gravity and separation to breasts, both anime and realistic, in situations where they are doing anything but. But to be clear — you probably don't need to use this LoRA unless your boob renders are going really wrong, or are just not the shape you were hoping for. Check the triggers for variation options. Breasts will get larger as you add LoRA weight. It may produce a bikini at a weight of 1.0 or higher, so use [bikini] style negative tags if you don't want that. To avoid this glitch, nude and nsfw were added to the token. It was trained on Chilloutmix but works well with all realistic models like <realvis20>
effect	breasts-bouncy		hentai boobs	Bouncing Breasts - Concept v1 by Moon_Light Recommended settings for getting good renders: # Weight ~0.8 # Works with most anime based model and not only # /parser:new /clipskip:2 (optional) Sidenotes: # Can easily handle rendering in resolutions such as 832x1024 (with <anylora> as base) # For extra bouncines include those prompts: bouncing breasts, unaligned breasts, motion blur
pose	breasts-grab-pov		boobs porn poses	POV Breast Grab by ekune
helper	breasts-helper		boobs anatomy	Breast-helper v1.1 by SysDeep Triggers: ## Breast Size ## a-cup breast b-cup breast c-cup breast d-cup breast ## Breast Size ## flat breast small breast medium breast big breast fake breast bimbo breast ## Breast form ## torpedo breast saggy breast ## Nipple Color ## pink nipples brown nipples ## Nipple Style ## normal nipples big nipples inverted nipples puffy nipples ## Areola Size ## small areola normal areola big areola
helper	breasts-nipples			masusu_BreastAndNipples by iawwa
clothing	breasts-no-bra		anatomy effects clothes porn hentai boobs	[LoRA] Braless by s4w3d0ff Braless as is wearing a shirt but without a bra under, so the breasts or nipples are covered
pose	breasts-on-glass		female poses environments	Reel BoobShowerPress by Reelai a Lora that produces photos of women with their breasts pressed against the shower glass Recommended settings for getting good renders: # Guidance at approx 7.0 # Weight is best between 0.8 and 1 # Works great with most realistic and un-realistic models # Works great with basic and elaborate prompts Sidenotes: # Reduce the weight to approx. 0.7 and use terms like, "bathtub, lying, bubblebath" to get some amazing bath scenes (use "ass" for a behind shot) # Seems to interact well with other loras including character loras (works great with <wet-hair>) # A good resolution to render at is /size:640x832 Basic Prompt: /render <breasts-on-glass>, breasts on glass, shower, 1girl, against glass, [blurry, lowres, worst quality] <deliberate2>
effect	breasts-perky		hentai porn boobs anatomy	Perky Breasts v1 by Gtonero Recommended settings for getting good renders: # To boost the effect add ‘perky breasts’ to your prompts. # Can be used with most anime base models and not only. # /parser:new /clipskip:2 (optional) Sidenotes: # Mixes really well with character LoRas # It’s easy to use and doesn’t require much tinkering.
effect	breasts-pressed		anime boobs female porn semi-real	Breast Press LoRA by DerHammer LoRA trained for generating the "breast-press" pose. Works well with anime and semi-real bases such as av5, revani. Trigger word is breast press, 2girls. You can add kissing and closed eyes / open eyes for those romantic moments. Add nipples to negative prompt to remove weird protrusions. A good weight range is from 0 - 0.8
helper	breasts-saggy		boobs anatomy	Sagging breasts by Gtonero LoRA to generate sagging breasts.
clothing	breasts-shirt-side		poses boobs porn clothes hentai	sideboob by aaamovaaa854 LoRA to generate sideboob renders.
clothing	breasts-shirt-under		effects boobs porn clothes hentai	Cropped shirt + underboob \|anime + realistic\|?LoRa? by dogu_cat LoRA to generate visible breasts under shirt.
clothing	breasts-shirt-under2		effects boobs porn clothes hentai	Upshirt / Underboob \| Clothing Lora 713 v1 by Numeratic LoRA to generate visible breasts under shirt. Works with anime and realistic renders.
pose	breasts-sucking		realistic semi-real poses boobs anime	No
helper	breasts-tiny		boobs anatomy	JK Small Breasts [Lite] by jk_chstr
helper	breasts-torpedo		porn boobs anatomy	JK Torpedo Breasts - Saggy boobs [EXPERIMENTAL] by jk_chstr
anime	breasts-ultra		boobs porn anime memes	boobcrush by throwawayjm
celebrity	brie		characters realistic female white	Brie Larson by losquit A model that resembles actress Brie Larson, created by LosQuit on Civitai. Suggested weight is 0.6 and some example base models that work well are <urpm> <aohard> and <chilloutmix>
effect	brightness		effects lighting	Brightness Tweaker LoRA (????LoRA) by ITvovTI Lora trained to adjust the brightness of your image. No tokens required. The recommended weight is 0.1 - 0.4, where higher weight is brighter the image. For negative weights, works well from -0.1 - 0.5 to make the render have a darker background, depending on the model
celebrity	britney-spears		characters female	This lora specializes in creating images of pop idol Britney Spears. It is intended to produce realistic and diverse portraits of her. The lora focuses on her visual appearance. For best results use a photorealistic model like: <photon>
clothing	bronze-punk		effects art style clothes fantasy armor	BronzepunkAI by konyconi Bronzepunk is a fantasy art style that tries to incorporate the bronze age armor fashion (primary) with modern elements of the steampunk aesthetic (lesser degree), making for some interesting gear and vehicles and machinery, especially. The creator, Konyconi, noted that you may need to lower your guidance if the faces glitch, due to so many trained pictures of persons wearing helmets. Try it with a weight of 0.5 and a base like <revani>
celebrity	brooke-shields		realistic female characters	Brooke Shields by garand343635 A realistic likeness of the model Brooke Shields. Defaults to her older years unless prompted otherwise. Try it with a realistic base model like <believe> or similar. This was trained on Cyber Realistic, and the creator doesn't recommend <realvis20> For the hazy photo studio glamour shot look, render at low guidance between 3.5-4.5
celebrity	brucelee		characters realistic male asian	Bruce Lee by 378866459393 An homage to the karate movie actor, Bruce Lee. A weight of 0.7 with most real style base models work well, like <realgalaxy>
celebrity	buffy		characters realistic female white	Buffy is a LoRA inspired by the look of Sarah Michelle Gellarr, aka Buffy Summers from Buffy the Vampire Slayer. A weight between 0.3 - 0.7 works best, with a real styled base model like <clarity>
lora	bukkake		porn hentai	Bukkake LoRA v1 by schneed It is rather prone to getting cum on hair, and has a slight bias against facials/cum on faces, so make sure to tune for that in your prompt so it gets proper coverage. It is able to generate excessive cum from a minimal 'bukkake, cum' tag alongside it. A strength of 0.75 is good. If your model still has a problem with handjobs, penises, etc. and the like, it'll only really make it work marginally better. Given that most models doesn't handle hands, penises, and other particulates regardless, expect only a marginal improvement. All v1.0 pics were made with my custom blends. All of these blends have an AOM3 base, so this LoRA should work for you as well if you use AOM3 or a similar model.
lora	bulge		porn penis memes	No
clothing	bunnygirl-anime		characters clothes anime	Bunny Girls v1.0 by nicerpanda # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low # Mangled limbs, general chaos ? Some LoRas can’t handle rendering with resolutions bigger than 512x512 // Also make sure your prompt isn’t too complicated for this LoRa # No visible effect ? Check if the LoRa has any triggers that need to be added in order to activate it # Good full models to try out LoRas with: anime ? <anylora> <counterfeit> <breakanime> comics ? <comics-vision> <icomix> <misstoon> semi-real ? <revani> <lyriel16> <dreamshaper6> realistic ? <majic> <bra5> <chilloutmix> <edge>
clothing	bunnygirl-costume		characters clothes realistic	No
clothing	burlesque		clothes	No
lora	burnin-kamiji-moe	No		No
pose	butt-feet		poses feet fetish porn	No
pose	butt-group-photo		butts poses	No
pose	butt-side		butts poses porn hentai	No
pose	buttjob		poses butts porn	No
pose	buttjob2		poses butts porn	How to use: # If you’re getting low quality renders, lower the weight. <buttjob2:0.25> is a good starting point to make it work. # In case of a ‘memory exceeded’ message, lower the number of steps, decrease the number of images & size, or try it with another base model. # It has problems with tendering in sizes bigger than 512x512. # It works best with anime based models, try it out with <anylora>.
lora	buttjob3			No
pose	buttlick		poses oral butts porn	Ass Licking model by Diedra, but there' more to it! The model was trained with square images with two women, so using it at 512x512 is best, but only one woman' face is likely if the photo is square. When you go wider, it might show both faces but the model maker says this is a higher chance for things to get janky. There' no extra triggers and we jam four words in the default token to make the effect work. Tested with <urpm>
pose	buttlick2		poses butts fetish porn oral	No
pose	buttlift		poses hentai butts porn	A lora for an ass support pose. It generates decent output, but some remixing might be needed to get better results. Recommended size - 512x768. Use with a weight between 0.4-1. It seems to work better with anime base models, but it’s possible to remix the generated pictures with a semi-real ones.
pose	buttpose		poses porn butts	The butt pose is a lora model called Butt Pussy by JakeRix. Try it with a weight around 0.5 - 0.6 with a real photo model like <chilloutmix>
style	calligraphy-art		art style effects tattoos detailers	Stable Diffusion is bad at text, and it' also bad at calligraphy... but this kind of calligraphy is more for abstract decorations. This model by Listy is intended to work with <lyriel16> and similar to create dark, dramatic images with calligraphy accents, more like a decorative art pattern. You can use this Lora to add intricate details to clothing, tattoos, and patterns on walls. If the effect is not what you want, try adding words like "lettering, text, calligraphy" to your prompt.
helper	cameltoe		anatomy porn	This LORA can be used to add the impression of a cameltoe. Start with low LORA weight, e.g. 0.3 and increase if the effect is not strong enough. Beyond a weight of 0.6 — any higher and it is very likely to give you only close-up panty shots. Works great with the base model <liberty>
lora	cammy-street-fighter	No		No
sci-fi	car-syd		vehicles sci-fi objects	Cardesigns by Syd Mead Style, inspired by the American Industrial Designer who passed in 2019 (rip). He was a neo-futurist designer best known for his work in the films Blade Runner, Aliens, Star Trek, and Tron. Interestingly, the model was trained on <chikmix> so use that as your base for best results, at a weight of around 0.6. Try wide aspect ratios as well as tall for different kinds of vehicle results
inversion	carla-gugino		female white characters	This textual inversion creates the likeness of film/TV star Carla G Recommended settings for getting good renders: # Guidance at approx 7.0+ # Best weight around 1.0 # Works well non-realistic and realistic models # works well with basic and elaborate prompts Sidenotes: # Works well with other loras # Render at low sizes and use /more /size /strength to upscale Example Prompt: /render 1girl, solo, carlagugino, <carla-gugino> [blurry, lowres, worst quality]<photon>
inversion	carlie-whalen		female characters realistic	This positive textual inversion creates images of social media influencer Carlie W Recommended settings for getting good renders: # Guidance at approx 7.0+ # Best weight around 1.0 # Works well non-realistic and realistic models # works well with basic and elaborate prompts Sidenotes: # Works well with other loras # Render at low sizes and use /more /size /strength to upscale Example Prompt: /render 1girl, solo,c4rl13wh4l3n , <carlie-whalen>,[lowres, blurry, worst quality] <cyberreal>
pose	carry-sex		poses porn	How to use: # It already contains most of the prompts needed for this particular position, so try it with something very simple first, # It has problems with tendering in sizes bigger than 512x512.
style	case		semi-real .base models semi-real art style	Knollingcase is famous for its iconic "case" designs, putting any object in a glass case of some kind. Insanely fine detailed people, objects, and landscapes inside glass cases, somewhat sci-fi. Incredibly unique! For something different, try a negative prompt to get rid of the box like [case, glass, box]
style	castlevania		anime cartoons art style gaming	The Castlevania Television Show Likeness LoRA by Lykon doesn't reproduce characters from the show by name, but you can describe their likeness with accurate prompting to get pretty close. It works at around a LoRA weight below 1 with base models like <anylora>
anime	cat-tail		furry anime anatomy	No
anime	cat-tail-fake		furry anime butts fetish bondage	No
lora	caught-concept	No		No
style	cell-shaded		art style effects	Designed by Thestanman on civitai. This lora is intended to give a cell shaded art style similar to telltale games, borderlands or breath of the wild. Lora works well at default weight or higher. Works best with realism models.
style	cetus		semi-real .base models anime fantasy	Like a mix between Revani + Lyriel, cetusMix is a pastel style anime model with detailed backgrounds that works well with anime LoRAs. The color pops off the page with the right prompt! Delivers amazing neon colors, and is really great for degen anime. Here' how to use it like a pro: https://www.animemaker.ai/cetus-mix-guide/
lora	cg-3d			3D CG Style Realistic by Jappww imitates the style of polished 3d renders, like Unreal Engine or Unity. It was designed to work with <chilloutmix> as the base, but may work well with other realistic base models.
anime	chainsaw-makima		characters female anime	No
lora	chainsaw-makima-real		female characters realistic semi-real	No
inversion	char-turner	No		No
pose	charactersheet		gaming illustrations poses anatomy anime characters comics 3d	Character Design Sheet (Helper) (3-perspectives) by YeiyeiArt Hello, guys! Today I bring you something special for all those who enjoy creating characters for their own projects. This model will allow you to create aesthetic perspectives to design your character in multiple views. Use the keyword to invoke the model: "CharacterSheet" for more precise results. Use the following words for better outcomes: "(multiple views, full body, upper body, reference sheet:1)". As an additional tip, the model works best with a resolution of 512 x 512 And the best weight "<0.5 - 0.8>".
lora	charlie-d'amelio		characters female realistic	No
pose	cheek-bulge		poses porn penis	Author states sometimes it can be too strong or weak depending on prompt complexity; ideal between 0.4 and 7. Bigger bulges with higher weights.
style	cheesedaddyland		semi-real art style environments illustrations paintings	CheeseDaddy Landscapes Mix is an extract of the larger CheeseDaddy base model, which is a fantasy model for semi-magical vast mountains, valleys, waterfalls, and other nature scenes. Crank it to a weight of 2 to get close to the base model' looks, or blend it with another base to land somewhere inbetween. A guidance lower than 7 is recommended overall
style	chikmix		semi-real .base models realistic asian korean	Chikmix is best described as a 2.5 Korean-style semi-realistic style, though some images can come close to real. Adding the triggers "digital illustration" or "realistic" can switch up the pain style. It is unfortunately a little difficult to work with, so keep your guidance under 10 to avoid artifacts
style	chilloutmix		realistic .base models semi-real asian	Chilloutmix is best known as an "asian beauty" ai model but it can also create other kinds of people, and is a popular LoRA base. We applied the latest fix for the "NI" version
inpainting	chilloutmix-inpainting	No		No
character	chinadoll		characters female asian realistic chinese	The likeness of a non-specific Chinese woman. Recommended settings: base: <chilloutmix> /sampler:dpm2m /steps:30 /clipskip:1 /guidence: between 5 to 9 LORA weight between 0.2 to 0.7
character	chinadoll2		asian female chinese characters anatomy	Chinese Doll Likeness 2.0 is a different, non-specific Chinese woman model by a different model maker of the other 'doll' creations. They recommend a weight of 0.5 and guidance of 7 with realistic base models like <chilloutmix> however in our testing this model resulted in a more semi-realistic look without strong negative prompts, so this may be one of the tricker in the 'doll' series. Some examples added glasses and red lipstick, though not official triggers of the model
character	chinadoll232		asian characters chinese female realistic	Version 4.2 of ZYDG323' Chinese Girl model is trained on 100 pictures of cute pictures of women, at 768x768. It will produce twins and defects at other resolutions. Works well with models like <chilloutmix> and <majic5> with a LoRA weight of 0.4 - 0.7
fantasy	chinese-dragon		fantasy chinese asian creatures	No
style	chinese-poster		chinese art style effects asian	No
clothing	chinesewedding		poses clothes chinese asian	Chinese Wedding is a lora for making two people stand in front of a red background in formal clothing. Naturally, it will make your characters Asian as you increase strength. Try it with a real photo model like <realvis13> and a weight of 0.8. Making the aspect ratio wider can help in adding more people to the photos. For semi-real paint effects, use <deliberate> as the base
celebrity	chloe-cherry		porn white characters female realistic	A likeness of Elise Jones, known professionally as Chloe Cherry, is an American actress, model, and former pornographic actress, created by model maker Rjox. Tested with <believe> as the base model, or similar
celebrity	christina-hendricks	No		No
lora	chrome-tan	No		No
sci-fi	chronomancy-mecha		anime mecha sci-fi fantasy armor comics	Chronomancy Mecha by Cinsdia Recommended for Mecha: anylora model, steps: 20, CFG scale: 7, clip skip: 2 No triggers needed! This model makes some cool looking Mechs, but don't pigeonhole it by thinking that is all it can do, It also makes some REALLY nice fantasy armor too (even better than Mecha). Check out this fantasy Mageward prompt from a lora user and prepare to be blown away... /render /size:512x768 /steps:30 /sampler:dpm2m /guidance:7 1 girl as a Mageward, best quality, masterpiece, ultra high res, (photo realistic:1.4), <chronomancy-mecha:0.9> [[easynegative, (nipple:1.4), logo, watermark, text, (worst quality:1.5), (low quality:1.5), (normal quality:1.5), low res, bad anatomy, bad hands, normal quality, black and white photos,(monochrome), (grayscale), cartoon,anime,drawing,painting,manga]] <revani>
sci-fi	citizen-starships		sci-fi environments objects spaceships	Scrap Shitizen \| Spaceships v1.0 by braintacles # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low # Mangled limbs, general chaos ? Some LoRas can’t handle rendering with resolutions bigger than 512x512 // Also make sure your prompt isn’t too complicated for this LoRa # No visible effect ? Check if the LoRa has any triggers that need to be added in order to activate it # Good full models to try out LoRas with: anime ? <anylora> <counterfeit> <breakanime> comics ? <comics-vision> <icomix> <misstoon> semi-real ? <revani> <lyriel16> <dreamshaper6> realistic ? <majic> <bra5> <chilloutmix> <edge>
anime	cjiang-cchan		characters female anime	No
style	clarity		semi-real realistic .base models fantasy	Clarity is a legit photo realism base model focused on anatomy, great for fantasy, and uses the "unstable diffusion" tagging system for ease of use. Great animals as well. This checkpoint does not contain any trigger words. However, placing some tags at the beginning of the prompts can heavily influence the generation. These tags include: "nsfw", "sfw", "erotica", and "nudity" \| "3d render", "cartoon" Note: For SFW generation, try adding sfw to your prompt and nsfw to your negative prompt. For NSFW generation, try adding either nsfw, erotica, or nudity to your prompt and sfw to your negative prompt. In general, this is more useful for generating sfw images. This concept also applies to 3rd render and cartoon. I recommend leaving 3rd render and cartoon both in your negative prompt for generating photographic images. Example prompt using commas and natural language: Positive A Professional Full Body Photo, of a beautiful young woman, clothed, standing indoors, Caucasian, toned physique, strawberry red hair, neutral expression Negative I recommend something simple like, deformed, bad anatomy, disfigured, missing limb, floating limbs, twisted, blurry, fused fingers, long neck, words, logo, text, mutated hands, mutated fingers Modify as needed. For example, adding 3d render, cartoon to your negative prompt will help generate photographic images. The prompts for this model are fairly flexible, experiment to find out what works best for you.
style	clay-render		art style effects	Clay Render LoRA model by LONGD, creates Unreal Engine style "clay" 3d figure that looks like something between an unpainted video game character or a flawless unpainted hyper detailed miniature statue. Tested with <revani> as a base and the results were stunning
lora	closeup-vaginal-creampie	No		No
clothing	clothes-explode		anime semi-real clothes effects	Pretty self explained. Makes clothes explode. Works best with semi-real and anime. I suggest sushi25 Pretty easy to get going, works around 0.1-0.3 the more weight the more exploded but it'll start deforming faces. Recommend using clipskip2. Example prompt: /render /sampler:dpm2m /guidance:8 /parser:new /clipskip:2 #fuji clothed alternative girl [monochrome, sepia] <clothes-explode:0.1> <sushi25>
inversion	clutter-home	No		No
inversion	clutter-mechanical	No		No
clothing	coat		objects clothes	A long non-specific kind of coat for anime and paintings, recommended lowering character weights under 0.5 weights and other light loras so it doesn't clash. Seems to work well under armor LoRAs as well, so try mixing and matching
lora	coco-bandicoot	No		No
style	cogecha-style		anime art style illustrations	No
clothing	collar		objects clothes objects bondage porn	A basic fashion or bdsm collar for role play, sometimes with a hoop in the front. For a larger one, try <collar2>
clothing	collar2		clothes objects bondage porn	A tall "posture" style collar for bondage. For a smaller one or fashionable look, use <collar> instead
style	coloring		.base models illustrations	Coloring Pages is a base model that creates black and white high-quality coloring pages from a text prompt. Use it to create a variety of subjects, like animals and landscapes. It does not require a complicated prompt to work. A guidance of around 8-11 and about 30 steps does the job well. In our testing we did see some colors and realistic photos come up, so if you run into this problem, try words like: coloring book ((illustration)), 2d, black and white, sharp focus, high contrast, [[[photograph, shading, shadows, pencil, gray]]]
styles	comi-noir2	No		No
style	comics		.base models comics cartoons illustrations	Comic Diffusion was the first really good breakout comics model for Stable Diffusion, released in November 2022. Trained on 6 styles at once, it allows anyone to create unique but consistent styles by mixing any number of the triggers. This is a full checkpoint base model, but it also has triggers!
lora	comics-2000		illustrations cartoons comics	No
style	comics-babes		anime .base models illustrations cartoons art style comics	When you mix the <babes> sultry model with the 2d comic book model <icomix>, you get the very fun Comic Babes versatile digital illustration checkpoint. Use this as a base model for beautiful hand drawn western style comics and cartoons. Adding "comic art" will punch the style bit to your prompts
lora	comics-bouncy		illustrations cartoons comics	General LoRa guide: # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low # Mangled limbs, general chaos ? Some LoRas can’t handle rendering with resolutions bigger than 512x512 // Also make sure your prompt isn’t too complicated for this LoRa # No visible effect ? Check if the LoRa has any triggers that need to be added in order to activate it # Good full models to try out LoRas with: anime ? <anylora> <counterfeit> <breakanime> comics ? <comics-vision> <icomix> <misstoon> semi-real ? <revani> <lyriel16> <dreamshaper6> realistic ? <majic> <bra5> <chilloutmix> <edge>
lora	comics-contact		illustrations cartoons comics	General LoRa guide: # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low # Mangled limbs, general chaos ? Some LoRas can’t handle rendering with resolutions bigger than 512x512 // Also make sure your prompt isn’t too complicated for this LoRa # No visible effect ? Check if the LoRa has any triggers that need to be added in order to activate it # Good full models to try out LoRas with: anime ? <anylora> <counterfeit> <breakanime> comics ? <comics-vision> <icomix> <misstoon> semi-real ? <revani> <lyriel16> <dreamshaper6> realistic ? <majic> <bra5> <chilloutmix> <edge>
effect	comics-cover		effects comics	No
lora	comics-dcau		illustrations cartoons comics	DCAU / DCAMU Style LoRA by Lykon “Trained mostly on Constantine movies and Young Justice. The goal was to make it able to do superheroes, but also magic/fantasy stuff as much as possible given the modern setting of the Constantine movies.” Recommended settings for getting good renders: # Raise weight to about 1.1-.1.2 in order to get more style out of this lora. # Try it out with <anylora> Sidenotes: # Easy to use.
lora	comics-dusk		illustrations cartoons comics	No
style	comics-jab		comics art style cartoons illustrations hentai	LoRa which art style resembles that of Jab Comix. To achieve the intended result it is recommended to use it together with <aohard> or <av45>. This LoRa can’t really handle rendering in sizes bigger than 640x640 and is oriented mainly towards NSFW content. Use with weight 0.5-0.75 and one of the triggers.
style	comics-jab2		comics cartoons art style	No
lora	comics-liu		illustrations cartoons comics asian	No
lora	comics-nutaku		illustrations cartoons comics	No
lora	comics-pear		illustrations cartoons comics	General LoRa guide: # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low # Mangled limbs, general chaos ? Some LoRas can’t handle rendering with resolutions bigger than 512x512 // Also make sure your prompt isn’t too complicated for this LoRa # No visible effect ? Check if the LoRa has any triggers that need to be added in order to activate it # Good full models to try out LoRas with: anime ? <anylora> <counterfeit> <breakanime> comics ? <comics-vision> <icomix> <misstoon> semi-real ? <revani> <lyriel16> <dreamshaper6> realistic ? <majic> <bra5> <chilloutmix> <edge>
style	comics-retro		art style retro comics	This LoRa will make your renders look more like an old-school comic, but it’s finicky and needs some tinkering. It works best with <misstoon> or <comicbabes>, but can be used with other base models. IMPORTANT - use it like this “<graphic-novel:0.1>”. Despite the low weight the effect will be visible, but raising it even to 0.2 will instantly lower the quality. There are no specific triggers or tokens required for this LoRa, but it seems to be good at producing neon sci-fi themed renders. It also produces interesting results with low guidance (~4-6).
lora	comics-sots		illustrations cartoons comics	No
clothing	comics-sunstone		clothes bondage fantasy comics cartoons	No
lora	comics-twisted		illustrations cartoons comics hentai	No
lora	comics-valerian		illustrations cartoons comics armor	No
style	comics-vision		.base models anime cartoons comics illustrations art style	Comics Vision v3 by Aledelpho Comics Vision (previously known as Toontanium) is a full checkpoint base model, a blend of models like MissToon and Icomix. As such, it serves as a good western-style illustration base model that' right on the edge of anime. The optional trigger Toontanium can alter the results, but is not included by default. For best results, render directly at 640x768 and then /facelift /anime with the ones you like. Here' prompting advice directly from the model' creator, Aledelpho: Start from the prompt of the pictures I've uploaded and then customize the areas focused on Character, Setting, Emotions, Colors and Aesthetic. For example: (inked original-art:1.2), (lineart:1.33),1woman,space-viking society, blonde braid strong manly woman barbarian,wrinkles,(depth),(movement:1.4),cinematic-composition,snob smirk,(hyperdefined),intricate-setting,undercut,yellow white orange,winter futuristic-city-streets,(thin-lines),viking-sci-fi aesthetic,neon-lighting,stars,giant-robots-machines,(fine-details:1.3),RAW-format,emotional,(muted-colors) Update v3: # To get crispier results use: /more /size:1200x1200 /images:4 /strength:0.45 on your favourite pictures. # Version 3 offers better quality and colour saturation while perfectly capturing the spirit of its predecessor.
object	condom-anime		porn hentai anime objects futa	How to use: # If you’re getting colorful noise renders, lower the weight. <condom-anime:0.1> is a good starting point to make it work. # Trained with Hassaku but works fine with most hentai models. # You don't need to use triggers but it does help. # Prompting a color works, e.g. white condom, green condom, pink condom etc.
object	condom-mouth		porn hentai anime objects	condom in mouth v1.0 by Hayasaka AI # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low # Mangled limbs, general chaos ? Some LoRas can’t handle rendering with resolutions bigger than 512x512 // Also make sure your prompt isn’t too complicated for this LoRa # No visible effect ? Check if the LoRa has any triggers that need to be added in order to activate it # Good full models to try out LoRas with: anime ? <anylora> <counterfeit> <breakanime> comics ? <comics-vision> <icomix> <misstoon> semi-real ? <revani> <lyriel16> <dreamshaper6> realistic ? <majic> <bra5> <chilloutmix> <edge>
anime	cop-solange		characters anime female	No
lora	cordelia-dead-estate	No		No
style	counter25		anime .base models environments illustrations	Counterfeit 2.5 is the older version of the model, which we liked for its detailed backgrounds and washed watercolor looks. The newer version of Counterfeit is 3.0, which has brighter colors and puts more emphasis on the subject. They're almost two completely different base anime models, and we love them both. Update: The colors are back. Now you can generate awesome backgrounds with this model while enjoying saturated colors.
style	counterfeit		.base models anime illustrations	Reinstalled 5.14 - The 3.0 version of Counterfeit has much brighter colors and more focus on the subject, whereas the previous model was more muted and pastel-like. The backgrounds look totally different in particular, so we're going to keep both. It should be noted that there are 10 upvotes and 3 comments about green spots from users on Civitai. We reinstalled with kl-f8-anime2 VAE
pose	covering-own-eyes		poses anime	An anime style pose of someone using one hand to cover their own eyes, typically smiling beneath. It' also possible to produce partial peeking
pose	cowgirl-asian		poses asian porn	No
clothing	cowgirl-bikini		clothes	Bikini Cowgirls by EDG
pose	cowgirl-pose		poses porn	No
pose	cowgirl-squatting-pov		poses porn	This is another simple sex position LoRA for POV Squatting Cowgirl. Author suggests using this LoRA at 0.9-1.0 weight for best results. Example prompt: /render <squatting-cowgirl-pov>, 1boy, penis, squatting cowgirl position, vaginal, pov If 'penis' is removed from the prompt it will give the effect of deeper penetration. If 'penis' and 'vaginal' are removed it will give the effect of her just sitting on you. If she is wearing a skirt or something similar it appear as implied sex - especially if you add 'cum'. If 'quatting' is added to the prompt it will show more of the position and a wider shot. This was trained on NAI model, iffy with more realistic models.
effect	crazy-expressions		effects expressions anime	Crazy Expressions supports 8 unique triggers to make it easier to control the facial expression of your subject. When using it, it is recommended to also turn on the "constricted pupils" followed by the facial expression for a more dramatic effect. The eyes may glow due to this model' training, for better or worse. A weight of 0.6 and anime base models like <av3> and <niji2> work well
lora	creampie-hairy		porn	Creampie & Hairy Pussy v1.1 by Wcxcxc169xoxo Suggest weighting below 0.7 and work with <chilloutmix> and ControlNet
pose	creampie-pov		poses porn butts	No
pose	creampie-pov2		porn poses butts	Close up Vaginal + Creampie LoRA v1 by KinkAI Author' comment: A LoRA for creating close up shots of vaginal sex in various poses and angles. Use the suggested trigger words to get the type of shot you want, They work in combination with each other, although some of them combine better than others. While testing I found that 'doggystyle from side' is least reliable, while 'doggystyle pov' and 'missionary pov' seem to work the best. I found that 'doggystyle from side' works better if you increase the strength of the LoRA. As always, you can use 'creampie' to add creampies to any of these. They tend to turn out quite intense, so if you're not into that turn down their strength by using (creampie:0.6) for example.
pose	creampie-standingvag		poses porn	This lora renders images of women being penetrated from behind (with optional creampie) Recommended settings for getting good renders: # Guidance at approx 7.0+ # Best weight around 0.8 - 1.0 # Works great with realistic and non-realistic models # works well with basic and elaborate prompts Sidenotes: # Seems to interact well with other loras # Render at low sizes and use /more /size /strength to upscale # The creampie is built in and can be a little over the top creamy. Use negative prompting to control. Example Prompt: /render <creampie-standingvag:0.8> standing vaginal close up, photo of 1girl getting fucked from behind by man in a office, (1girl), (1boy) [lowres, worst quality, blurry] <photon>
lora	creatures		creatures fantasy	Fantz Creatures Design is the more flexible full body and portraits Lora, meant to support fantasy monster and wild beast creations. Try it with fantasy models like <atoz> at a weight of 1 or less
sci-fi	creatures-portraits		creatures fantasy sci-fi	Another creatures Lora, this one specifically for close-up images in the Fantz art style. If you need a shot with the legs, use the other <creatures> Lora instead. A weight of 0.9 with any realistic or semi-real model like <realvis13> works well
style	creepy		realistic .base models fantasy art style	Creepy Diffusion is an older Stable Diffusion low resolution model trained on Internet Folklore images of decrepit hallways and monsters of dubious quality. The photos are messy and unclear, but that' exactly the point. This is a base model, so you can boost the reality by mixing it with LoRAs
lora	croissantify	No		No
lora	croptops			Crop Tops by fitCorder on Civitai will dress your models with a bandeau bra or bare shoulders top depending on the prompt instructions. The prompt must include the trigger word 'bandeau' and can be mixed with other descriptors such as small top, Crop Top, tank top, Slingshot Swimsuit, bikini, etc.
clothing	cross-dressing-man		lgbtq male clothes characters	No
effect	cum		porn effects	Instant Cumshot or simply <cum> is the April 29th update to <instajizz>. This one seems to affect the facial expression more than the prior. They are the same LoRA by the same creator, but somehow have different effects, so try both! Check triggers for more options. Notes from the creator: It seems most useful at around 0.4 LoRA strength, which covers faces etc. in cum, but around 0.9 strength is more likely to actually make an action shot with a dick firing off a volley. Beware, in-between numbers like 0.7 generated some floating dickblobs in my testing! Avoid the dickblob valley. Trained entirely on images of professional adult actresses, as freeze frames from 1080p+ video. About half are POV kneeling shots, with some side view or laying down or multiple women or multiple men, and about all are frames where there' actually a big rope of cum shooting out of a dick. Surprisingly flexible given the limited variety of inputs. While you can replace "cum on face" with "cum on stomach" etc., the scale of the cum ropes stays the same and the realism drops off when the subject is zoomed out. There' probably some way around this. Hentai checkpoints respond fine to the trigger words without loading this LoRA, but it does seem to help achieve a targeted, ropey look in models that tend toward the drips all over look
effect	cum-bath		porn effects	No
pose	cum-facial		female porn poses	a Lora that produces photos of facial cumshots Recommended settings for getting good renders: # Guidance at approx 5.0 - 7.0 # Weight is best between 0.8 and 1 # Works great with most realistic and un-realistic models # Works great with basic and elaborate prompts Sidenotes: # Seems to interact well with other loras including character loras (works great with other oral-based loras like <bj-pov>) # A good resolution to render at is /size:640x640 & /size:640x832 # Less weight gives less cum Basic Prompt: /render 1girl, solo, facial, cum, <cum-facial>, [lowres, blurry, worst quality] <deliberate2>
pose	cum-facial-mask		oral porn fetish poses	No
pose	cum-in-mouth		poses effects porn	No
inversion	cum-in-mouth-ti			No
lora	cum-in-nose		porn fetish	How to use: # If you’re getting red noise renders, lower the weight. <cum-in-nose:0.1> is a good starting point to make it work. # In case of a ‘memory exceeded’ message, lower the number of steps, decrease the number of images & size, or try it with another base model.
effect	cum-on-tongue		effects porn oral	# If you’re getting colorful noise renders, lower the weight. <cum-on-tongue:0.1> is a good starting point to make it work.
pose	cum-real		porn poses female	This lora produces female faces covered in realistic cum. It works great in realistic models and also seems to work well with anime-based models. Keep the weight at 0.6 and above or things tend to get funky. I'd recommend 0.9-1 for best results. Example prompt: /render photo of 1girl, solo, a beautiful woman, on the beach <cum-real> sperm on her face, outdoors, sunny, [blurry, lowres, worst quality] <realvis20>
lora	cumflation	No		No
pose	cunnilingus-pov		porn oral pussy poses	How to use: # Hard to use. # Try it out with weight 0.3-0.6 # It has problems with tendering in sizes bigger than 512x512. # It’s quite picky when it comes to base models, but <aohard> works.
clothing	cute-dresses		clothes	No
lora	cutegirlmix		realistic anatomy female chinese asian	A general booster for beautiful women trained on Chinese TikTokers, though doesn't generate a specific person per se. Try it with a base like <chilloutmix> at a weight of 0.4 - 0.7
sci-fi	cyberdead		effects horror sci-fi	It generates cyborg zombies, the nsfw kind! Try it with <revani> as your base, and a lora weight of 0.75 with words like grotesque and any nsfw body part
sci-fi	cyberpunk-edgerunners		anime art style sci-fi	Cyberpunk Edgerunners LoRa is focused on bringing this particular style from anime into your renders. # Works great with any weight. You can adjust it accordingly. # Can render at sizes such as 832x1024. # Good results with most base anime models. Example prompts: cyberpunk edgerunners, 1girl, animal ears, blonde hair, bob cut, colored sclera, cyborg, fox ears, fox girl, fox tail, japanese clothes, kemonomimi mode, kimono, kiseru, looking at viewer, mask, mouth mask, obi, red eyes, red kimono, sash, science fiction, short hair, smoke, smoking, smoking pipe, solo, tail, upper body, yellow sclera, ((masterpiece)), [painting by bad-artist-anime, painting by bad-artist, watermark, text, error, blurry, jpeg artifacts, cropped, worst quality, low quality, normal quality, jpeg artifacts, signature, watermark, username, artist name, worst quality, ow quality, bad anatomy]
anime	cyberpunk-lucy		characters anime female sci-fi	Cyberpunk Lucy LoRa is focused on this particular character from the anime. This LoRA is flexible enough to allow you to change her hairstyle and clothes. If you type jacket, you get her jacket, and if you type shorts, you get her shorts, etc. # Works great with any weight. You can adjust it accordingly. # Can render at sizes such as 832x1024. # Good results with most base anime models. Example prompts: lucy \(cyberpunk\), 1girl, against railing, arm rest, bangs, bare shoulders, belt, black belt, black leotard, black pants, blurry, bob cut, breasts, building, cigarette, cityscape, clothing cutout, cropped jacket, cyberpunk, depth of field, from side, gradient eyes, grey eyes, grey hair, holding, holding cigarette, jacket, leotard, lips, long sleeves, looking afar, looking ahead, mechanical parts, medium breasts, multicolored eyes, multicolored hair, night, night sky, off shoulder, open clothes, open jacket, outdoors, pants, parted lips, railing, red eyeliner, science fiction, short hair with long locks, short shorts, shorts, sidelocks, sky, smoke, smoking, solo, standing, teeth, thigh cutout, upper teeth only, white jacket, white shorts, cyberpunk \(series\), cyberpunk edgerunners [painting by bad-artist-anime, painting by bad-artist, watermark, text, error, blurry, jpeg artifacts, cropped, worst quality, low quality, normal quality, jpeg artifacts, signature, watermark, username, artist name, worst quality, low quality, bad anatomy, watermark, signature, text, logo]
anime	cyberpunk-rebecca1		characters anime female sci-fi	This LoRa is focused on this particular character from the anime. Compared to the other Rebecca LoRa it can reproduce her much more faithfully. hairstyle (swap) twintails, black hairband <-> long hair jacket (swap) black jacket <-> sleeveless mechanical Hand (Quality is not guaranteed) mechanical hands # Works great with any weight. You can adjust it accordingly. # Can render at sizes such as 832x1024. # Good results with most base anime models and some real ones. Example prompts: masterpiece, best quality, highres, 1girl, rebecca \(cyberpunk\), solo, twin tail, hairband, mechanical eye, colored sclera, red sclera, colored skin, white skin, leg tattoo, neck tattoo, green hair, long hair, small breasts, underwear, fang, red pupils, skin fang, red eyes, black jacket, hand on hip, flower field, leaning forward, evil grin, [lowres, blurry, worst quality, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, low quality, normal quality, jpeg artifacts, signature, watermark, username, blurry, lowres graffiti, low quality lowres simple background]
anime	cyberpunk-rebecca2		characters anime female sci-fi	This LoRa is focused on this particular character from the anime. Compared to the other Rebecca LoRa it produces a more mature version of Rebecca. # Works great with any weight. You can adjust it accordingly. # Can render at sizes such as 832x1024. # Good results with most base anime models and some real ones. Example prompts: masterpiece, best quality, highres, 1girl, rebecca \(cyberpunk\), solo, twintails, hairband, mechanical eye, colored sclera, colored skin, white skin, leg tattoo, neck tattoo, green hair, long hair, small breasts, black bra, thong, fang, red pupils, skin fang, red eyes, sleeveless, indoors, standing, arm behind head, hand on hip, cowboy shot, smile, open mouth, [lowres, blurry, worst quality, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, low quality, normal quality, jpeg artifacts, signature, watermark, username, blurry, lowres graffiti, low quality lowres simple background]
style	cyberreal		realistic semi-real .base models female	CyberRealistic v3.1 by Cyberdelia Cyber Realistic is a base model focused on female portraits, specific to LoRA use. It' suitable for realistic and semi-realistic work. Despite the cyber name, it isn't focused on sci-fi but it can do cyberpunk scenes well enough. Recommended for beginners, it can do a lot with a little. One of the model' key strengths lies in its ability to effectively process textual inversions and LORA, providing accurate and detailed outputs. Additionally, the model requires minimal prompts, making it incredibly user-friendly and accessible.
inversion	cyberrealistic-negative		.negatives	CyberRealistic Negative by CyberDelia is a companion designed for use alongside the <cyberreal> base model. Usage: [[<cyberreal:-1>]]
celebrity	daddario		female characters realistic	No
celebrity	daenerys		characters female fantasy semi-real white	A portrait of a pretty blonde woman resembling Khalisi from Game of Thrones, by model maker BenyGoren. Try with semi-real and base at a weight of 0.7 like <dreamshaper> for the intended effect. The model is a bit limited in its poses and requires /guidance:7 or lower To get a realistic version, try something like /render /guidance:7 a realistic portrait, photo of princess halisi daenerys from game of thrones in a castle <daenerys:0.4><urpm13>
celebrity	dafoe		characters realistic male white	Get the thin body and vascular arms you've always wanted in your manly AI art by this gorgeous sextacular Lora by the world' most handsome man, but not exactly him. It' inspired by actor Willem Dafoe. A recommended weight of 0.9 or lower is best. Works with <realvis13> and similar at a guidance of 6
sci-fi	daisy-ridley		characters female realistic sci-fi star wars	No
celebrity	dakota-fanning		characters female white	No
effect	danni		characters effects semi-real female	Danni is a blonde large breasted porn star character inspired by Danni Ashe. Works with a variety of base models at a weight between 0.5 - 0.7, with the lower end being necessary if you stack other LoRAs
lora	danni2		characters female porn boobs white	A newer version of the Danni Ashe pinup likeness by a different creator, this very greatly improves facial likeness without complicated prompting. Remember to mix with an appropriate base model for best results, like URPM 1.3, though semi-real base models also look cool for a different style
celebrity	daphne-blake		cartoons characters	Recommended settings for getting good renders: # Guidance at approx 7.0+ # Best weight right around 0.3 - 0.6 # Works best with non-realistic models eg. <atoz>, <toonyou> # Works great with basic and elaborate prompt # Use /recipe:easynegative when possible Sidenotes: # Seems to interact well with other loras # Render at low sizes and use /more /size /strength to upscale # There are some great opportunities here to remix your final renders with a realistic model like <level4> to get some great results! Example Prompt: /render /recipe:easynegative 1girl, solo, cartoon style, daphneblake, <daphne-blake:0.5> [blurry, lowres, worst quality]<toonyou>
effect	dark-and-light		effects art style lighting environments	A dramatic dark and light effect with a high-contrast illumination surrounded by darkness. Hires fix is recommended with a recommended weight of 0.7-1.0. Works well with anime base models. A little fussy
lora	dark-and-light-v2			No
style	darkestdungeon		gaming art style environments	V1.1 An experimental dungeon environment fantasy lora based on the videogame Darkest Dungeon, though a bit more anime styled as the recommend base is <revani>
anime	darkmage-girl		characters anime female	The Dark Magician Lora by Lykon. Should work on most anime models at 0.6-0.7 weight, as well as semi-real models as well as <chilloutmix>
style	darkrev		.base models semi-real anime fantasy	DarkRevPikas v2.0 by BixBit11 This model is a merge of three interesting checkpoints: Dark Sushi Mix <sushimix>, ReV Animated <revani>, Pika' Animated Mix. Thanks to this it can generate a variety of different styles and situations. Tips: # It' good at generating images with sizes such as 640x832 or 832x640. # Try it out with clipskip:2. # To get more dimensions with 2.5 images, I recommend treating them with this command: "/more /size:1400x1400 /strength:0.4 /guidance:15 /images:2 <darkrev>" But only 2.5D! It will ruin the art style of 2D anime ones. # For a more realistic image use: photorealistic, (hyperrealistic:1.2), beautiful, masterpiece, best quality, extremely detailed face, perfect lighting, For less realism remove photorealistic and (hyperrealistic:1.2) # Good negative prompts: (worst quality, low quality:1.4), (monochrome), zombie, watermark, username, patreon username, patreon logo,
style	darkvoyage		realistic .base models semi-real fantasy	Dark Voyage is an early experimental model trained on a small number of Midjourney 3 images. The images are aesthetically pleasing compositions, but it tends to want to do its own thing because of it' limited scope. Try making desert Cronenberg movies, garden gnomes, machines, and mechanical stuff with it at a guidance of about 3.5. It' a base model, so it works with other LoRAs
clothing	darthvader		characters sci-fi clothes star wars armor	An interesting character LoRa that can be used for more than just what its name suggests. When used with low weight (0.1-0.6), it will first try to change the character' clothes without affecting the gender. Thanks to this, it can be easily mixed with many other LoRas making the resulting renders look darker and more sci-fi. Of course, reinforcing it with relevant prompts and cranking the weight up to 0.8-1.2 will produce an expected result - the dark lord himself.
celebrity	dave-foley	No		No
anime	dbz-bulma		characters anime female retro	An anime character that looks like Bulma from Dragon Ball Z, created by model maker Kirazu. Use a weight of 0.7 and add words like [figurine] if you get toy-like aesthetics. A good base is <aom3>
anime	dbz-bulma2		female characters anime dragonball	No
inversion	deep-negative	No		No
base	degen		realistic .base models semi-real fantasy porn	Degenerate is a base model checkpoint blend of other models including URPM, Chickmix, Chillout, and AOM3 among others. It' good for beginners as it emphasizes compatibility and does well with low steps and guidance. If you like URPM, this is a great alternate to change things up. It' best for real, and stylized semi-real images
style	deliberate		semi-real .base models realistic	A catch-all for Deliberate2
base	deliberate2		semi-real .base models realistic realistic environments interiors architecture	Deliberate 2 has been called "the NSFW master' toolkit", boasting the best balance of detailed backgrounds, light effects, skin texture, and accurate prompts at the time of its release. Incredibly versatile, though prompts tend to run on the long side and therefor isn't recommended for beginners (use Level4 instead). Works very well as a semi-real base for LoRAs. You can get photorealism out of it, but it' going to make you work for it
inpainting	deliberate2-inpainting	No		No
style	design-y2k		art style memes	No
lora	detail-enhancer		detailers	No
effect	detail-make		effects realistic semi-real detailers	LoRA Name: Tangbohu Detail Maker LoRA (???????LoRA) Lora is used for adding details. Lora is updated to ver 2.5. Weight range is from -6 to 3, applies for anime, semi-real, real bases. Positive weights increases detail of image, allowing making details out of air. Negative weights can be used to reduce details. 2D or 2.5D models will recreate 3D realistic effects when using negative weights. Experiment with 0.5 steps still you get the rough amount of detail you intend, then change by 0.1 steps.
inversion	detail-plusplus		detailers	No
lora	detail-simplify		detailers	BD Marumi LoRA v.3 by br d A lora for adding or removing extra details to your renders. Based on breakdomain model series. Using it with minus weight like <break-simplify:-1> will lead to more detailed outcome, while doing the opposite will provide more simplified renders. For stronger effect use high weights like: <break-simplify:1.5> <break-simplify:-1.5>
effect	detail-tweak1		anime semi-real realistic detailers effects	LoRA designed for removing or adding detail while keeping the style of the render the same. Current strength is good from 0 - 2 depending on your use case. In the future, it can be used as negative weight to reduce details. Works with anime, semi-real and realistic models and styles.
effect	detail-tweak2		anime semi-real realistic detailers effects	No
sci-fi	diesel-punk		semi-real art style sci-fi environments armor	A LoRa that will fill your renders with retro-futuristic machines and military uniforms. Try it out with fantasy and semi-real models like <revani>. Easy to use, but keep guidance below 9 to avoid artifacts and distortions. It works best with a weight between 0.8-1.1, and doesn’t require complicated prompts to produce good results.
pose	dildo		objects hentai hentai porn	FunToys by Yatsukiko is a miscellaneous sex toy Lora that mostly creates dildos. This creator dumped the contents of their training data, so here we can study which tags work best with the model. They note that "xray" is not one of them. https://yatsukiko.me/funtoys/ Weight: 0.5 - 0.6 works best. Other trigger words: Vaginal stuff: vaginal_masturbation, dildo_in_vagina, vaginal, vagina, vaginal_inserition, vaginal_object_insertion Anal stuff: anal, anal_insertion, anal_object_insertion, anal_masturbation, anal_beads. Futanari: It somewhat works, but you may need to add more weight to it. Also, if the sex toy is a dildo, it will most likely have a penis texture. You may want to add some other models for better futa results. Base model recommendations: chikmix, revani
lora	dilraba		asian chinese characters female	Dilraba Dilmurat, is a Chinese actress, host, dancer, singer and model of Uyghur ethnicity, known for Eternal Love of Dream (2020), Eternal Love (2017) and The Flame' Daughter (2018)
lora	discord-chan			characters, female, anime, memes
anime	disgusted-look		expressions anime	This LoRa does exactly what its name says. It works best with anime models but may produce funny results with others. Weight can be adjusted according to preferences, but the higher it is, the more this LoRa tends to focus on the face only. There’s no need to add any extra prompts as they’re already included in tokens. It has a tendency to generate NSFW content.
lora	distance-control			/concept /description:Distance Control by YouKengi can support deeper backgrounds and character position. It was designed for use with an anime base model, so usage with realistic models may encounter some unwanted illustrated style transference. Usage: A negative weight like <distance-control:-2> creates a distant focal point and the character will appear small and far away from the camera. A positive weight like <distance-control:2> will bring the subject much closer to the camera, and the background may lose depth, such as a street or alley may vanish. The creator also recommends starting your prompt with /parser:new /clipskip:2 to improve the quality The LoRA has various limitations depending on the prompt. For example, if the majority of the prompt is about the character and it is zoomed out, you may need other supporting LoRAs. The creator recommends pairing with <background-details>
anime	dizzy-guilty-gear		characters female gaming anime	No
lora	doa-kasumi		gaming female characters	A likeness of Dead or Alive' Kasumi video game character. Tested and working with the base model <believe>
pose	doggystyle-side1		poses hentai porn butts	No
pose	doggystyle-side2		poses hentai porn butts	No
pose	doggystyle-side3		poses hentai porn butts	No
pose	doggystyle-tiptoe		poses porn hentai	No
pose	doggystyle-tree		butts hentai porn poses	This LoRa is really unstable; the intended effect appears only sometimes. Add all triggers and some more, then try rendering at 512x512. However, when used without triggers, it seems to improve the color saturation of some models.
pose	doggystyle1-pov		poses hentai butts porn	No
pose	doggystyle2-pov		poses porn	No
lora	doja-cat	No		No
lora	doodle-style		art styles anime illustrations	No
pose	downblouse		poses hentai anime	No
clothing	downblouse2		poses clothes porn boobs	A clothing and pose modifier for anime and beyond, this model is called "Extended" Down Blouse, referring to a bigger range of flirting or peeping
style	dragon-age		anime cartoons art style	Dragon Age Absolution LoRA by Lykon doesn't reproduce characters from the show by name, but you can describe their likeness with accurate prompting to get pretty close. It works at around a LoRA weight below 1 with base models like <anylora>
sci-fi	dragon-scales		effects art style fantasy sci-fi	No
style	dreamful3		.base models art style illustrations semi-real	Dreamful Version 3 by Hius is a semi-realistic model, but aims to create the most realistic illustrations through natural lighting and 3D simulated effects. Use it to create fantasy art, illustrations, 3d paintings, and Unreal Engine style game characters. This is a new base model, so probably not as compatible with your favorite LoRAs as others.
style	dreamlike		semi-real .base models fantasy art style interiors architecture	Dreamlike is an older but solid illustration and semi-real base model, best for playful colorful images like stock photos and concept art. Dreamlike has an incredible range and is easy to work with, one of the better choices for beginners creating fantasy and semi-real works.
inpainting	dreamlike-inpainting	No		No
style	dreamshaper		semi-real .base models fantasy environments interiors architecture	Version 6 of Dreamshaper is a high-quality semi-real base model, now more realistic and has noise offset for deeper contrast. A good base choice for fantasy renders in general. The earlier versions leaned more towards anime, whereas this version is more semi-realistic. It' less likely than many models to create NSFW, and is harder to prompt for, so not recommended for beginners. Try the #fantasy recipe for portraits using <dreamshaper> While it generates very similar renders to V5, this version tends to add more details.
style	dreamshaper7		.base models fantasy realistic semi-real	An update to the base model Dreamshaper by Lykon, an update to the popular choice for fantasy art creators. Version improves LoRA support, NSFW and realism. Suggested settings from the creator: Sometimes /clipskip:2 helps, always use /highdef, don't use /facelift as it will wash out the paint effects
inpainting	dreamshaper7-inpainting	No		No
style	dreamshaper8		semi-real fantasy paintings .base models	Dreamshaper 8 by Lykon is one of the most popular base models for illustrations and art. This update includes better support for character LoRAs, better photorealism, can still handle anime, and better NSFW
lora	drive-thru		environments memes	A drive-thru teller window for fast food. The tokens include the drive-thru costume and microphone hardware for your convenience
anime	drow		anime fantasy characters	Drow Concept LoRa by lykon Whether you are trying to make Drizzt Do-Urdon, an anime Dark Elf, or one from World of Warcraft; this Lora will fit that need. Trained on DreamShaper 7, triggers with drow, plus other tokens such as colored skin, dark elf, blue skin, grey skin, pointy ears. Should work on basically any model and any style, but I tested it on AAM, AnyLoRA, DreamShaper and AbsoluteReality. Should work around weight 1.
style	dublex		.base models art style environments semi-real	Double Exposure Diffusion by Joachim Sallstrom. This is the original base model, Version 1, for creating ephemeral double exposure images. It was reinstalled to optimize our network for more reliable delivery of this classic model
style	duchaiten		semi-real .base models fantasy architecture interiors	Duc Haiten' AI art model received praise for handling a more diverse set of characters and 18+ situations. It is a 3D style semi-realistic fantasy model with emphasis on dramatic lighting and portrait composition. Also great for semi-realistic anime
style	ducjourney		semi-real .base models fantasy art style	DucHaitenJourney v5.6 by DucHaiten Artist Duc Haiten' take on the midjourney style. Has deeper tones and is more stylized towards realism. V5 brings the following upgrades: "Images are sharper, colors and lighting continue to be improved to be more realistic, Asian characters look better, images are more detailed, prompts are more diverse, the face looks more natural, the expression is more soulful, and finally reduced anatomical error." Suggested negative prompt: illustration, 3d, 2d, painting, cartoons, sketch, (worst quality:1.9), (low quality:1.9), (normal quality:1.9), lowres, bad anatomy, bad hands, vaginas in breasts, ((monochrome)), ((grayscale)), collapsed eyeshadow, multiple eyebrow, (cropped), oversaturated, extra limb, missing limbs, deformed hands, long neck, long body, imperfect, (bad hands), signature, watermark, username, artist name, conjoined fingers, deformed fingers, ugly eyes, imperfect eyes, skewed eyes, unnatural face, unnatural body, error, bad image, bad photo
lora	DudouTop		clothes anime semi-real realistic realistc	The dudou is a traditional Chinese underwear. In Western fashion, it is sold as a sleeveless shirt and backless halter-top blouse. Perfect for Summer, the dudou top unveils the shoulders and back while genlty espousing the natural curves of the woman' chest. This lora produces the Dudou top and matching panties. Use it in your prompt with a weight of 0.6: (white DudouTop), (white Panties) <lora:DudouTop:0.6> and mix it with different styles between the brackets: see-though or lace or silk or satin in the color of your choice Very nice work by SettingsNoProblem on Civitai. This alluring top an be used in renders with anime, semi-realistc and realistc models.
sci-fi	dune-style		environments sci-fi fantasy	Introducing Dune Style Model: Embracing the Aesthetics of Dune (2021) Dune Style Model, driven for 10 epochs and 7940 steps, embodies the captivating aesthetics of the 2021 movie Dune. Inspired by the visual style of Dune, the model excels at generating images that capture the atmospheric desert landscapes, intricate set designs, and richly detailed costumes that define the film.
style	e18		semi-real .base models illustrations furry	No
pose	eager-pet		poses hentai fetish	Eager Pet Pose v1.0 by pensive_penguin # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low # Mangled limbs, general chaos ? Some LoRas can’t handle rendering with resolutions bigger than 512x512 // Also make sure your prompt isn’t too complicated for this LoRa # No visible effect ? Check if the LoRa has any triggers that need to be added in order to activate it # Good full models to try out LoRas with: anime ? <anylora> <counterfeit> <breakanime> comics ? <comics-vision> <icomix> <misstoon> semi-real ? <revani> <lyriel16> <dreamshaper6> realistic ? <majic> <bra5> <chilloutmix> <edge>
anime	earth-chan		characters female anime	No
inversion	easy-negative	No		No
sci-fi	ebonygoldmecha		anime mecha sci-fi fantasy armor comics	Ebony Gold Mecha by Cinsdia Recommended for Mecha: anylora model, steps: 20, CFG scale: 7, clip skip: 2 No triggers needed! This model makes some cool looking Mechs, but don't pigeonhole it by thinking that is all it can do, It also makes some REALLY nice fantasy armor too (even better than Mecha).
lora	eddie-murphy		characters realistic male black	No
base	edge		realistic .base models semi-real realistic	Edge of Realism from the maker of Level4. A new photorealistic model. The base is Level 4 (v5, which we also host) but using high quality original photos for a completely different look. Easy to prompt for, good for beginners. A nice base model for portraits
lora	eida-boruto	No		No
style	eldenring		gaming .base models fantasy	No
clothing	elegant-dresses		clothes	No
lora	ellie		characters female realistic gaming semi-real white	No
lora	ellie-tlou		characters female gaming	This lora renders images of Ellie from The Last Of Us (game) Recommended settings for getting good renders: # Guidance at approx 7.0+ # Best weight around 0.15 - 0.2 # Works great with realistic and non-realistic models # works well with basic and elaborate prompts Sidenotes: # Seems to interact well with other loras # Render at low sizes and use /more /size /strength to upscale # for more real world renders use negative prompts like: dirt, bruises, backpack Example Prompt: /render 1girl, solo, <ellie-tlou:0.175>, [worst quality, lowres, blurry] <cyberreal>
lora	ellie-tlou-2	No		No
anime	elphelt-valentine		characters female gaming anime	No
lora	elsa-jean		characters female	Sapphire Nicole Howell (born September 1st, 1996 in Canton, Ohio) is an American pornographic actress. Her acting name, Elsa Jean, is taken from the character of Elsa, the princess heroine of the animated film Frozen.
lora	elysia		characters anime female	No
lora	emi-takei		characters female japanese asian	No
lora	emiru-twitch		characters female realistic	No
lora	emma-stone		characters female white	No
lora	emma-watson		characters realistic female white	No
inversion	emma-watson-art		characters female paintings semi-real realistic	Emma Watson by Dogu Cat is the more 'artistic' inclined emma model, example prompt: <emma-watson-art> in full steel knight armor, highly detailed, heart professional majestic oil painting by Vasily Surikov, Victor Vasnetsov, Konstantin Makovsky, trending on ArtStation, trending on CGSociety, Intricate, High Detail, Sharp focus, dramatic, photorealistic [[tan skin, old, bored photo,no color, blurry, ugly, poor quality, deformed hands, deformed face, deformed body, extra limbs, low quality, normal quality, text, errors, bad anatomy, mutation, deformed fingers, missing fingers, low res, bad hands, cropped, deformed hands, deformed legs, deformed arms, multiple arms, signature,Scribbles,Low quality,Low rated,Mediocre,3D <deliberate2>
lora	emma-watson-face	No		No
inversion	emma-watson2		characters realistic female	Emma Watson V2 by JernauGurgeh. Recommended prompt: <emma-watson2>, a woman as a movie star, modelshoot style, (extremely detailed CG unity 8k wallpaper), photo of the most beautiful artwork in the world, professional majestic oil painting by Ed Blinkey, Atey Ghailan, Studio Ghibli, by Jeremy Mann, Greg Manchess, Antonio Moro, trending on ArtStation, trending on CGSociety, Intricate, High Detail, Sharp focus, dramatic, photorealistic painting art by midjourney and greg rutkowski, (white turtleneck top:1.2), ((movie premiere)), (long skirt), ((standing near a movie theater)), ((paparazzi in the background)), (looking at viewer:1.2), (detailed pupils:1.3), ((closeup portrait:1.1)) <believe>
clothing	emo		memes clothes expressions	That 2005-2010 emo "light goth" fashion aesthetic style: Poofy hair, eyeliner, emo, emo makeup, goth. Good LoRA strength is between 0.3-1.0 depending on the model. Try <chilloutmix> as the base
pose	emotion-puppeteer		characters poses expressions male female	a Lora that produces dynamic emotional expressions for your characters Recommended settings for getting good renders: # Guidance at approx 5.0 - 7.0 # Weight is best right at 1 # Works great with realistic models and non-realistic models # Works great with basic and elaborate prompts ***To get what you want, you should read this. Use this prompt format for best results "{A} face, {B} eyes, {C} mouth, {D}, {anything else}" e.g. "plain face, seething eyes, screaming mouth, closeup portrait, looking away, old man with white hair" CFG = 5 is recommended A - Face keywords plain, unattractive face - my subjective opinion - biased towards extreme emotions cute, attractive face - my subjective opinion - biased towards pleasant emotions adorable, most attractive face - my subjective opinion - strongly biased towards pleasant emotions B - Eye keywords seething, >: shaped brows - scrunched brows grieving, <: shaped brows - scrunched brows widening, eyelids raised very high - brows arched glaring, lowered brows - eyelids lowered pleasing, neutral brows - eyes crecent shaped and smile crinkled piercing, neutral brows - eyelids neutral squinting, lowered brows - eyelids squeezed closed or almost closed winking, right eye open, left eye closes C - Mouth keywords breathing, mouth relaxed and slightly open resting, neutral or slight relaxed frown - lips together smiling, smiling with lips parted - teeth together or slightly apart screaming, mouth wide open - lower-teeth visible or chin scrunched sneering, nose scrunched - left upper-lip raised - upper-teeth not bared singing, mouth wide open and oval-shaped delighting, smiling with mouth wide open kissing, puckered lips gnashing, nose scrunched - upper-lip evenly raised - teeth together - lower-teeth bared gasping, mouth open and frowning-bean-shaped - chin relaxed relaxing, slight relaxed smile - lips together frowning, chin and or nose scrunched - lips together snarling, nose scrunched - upper-lip evenly raised - lower-teeth not bared pouting, slight frown - lower-lip extended grimacing, lips apart and stretched wide - lips apart - teeth may be visible D - Other keywords extreme closeup portrait, cropped at neck or closer closeup portrait, cropped at shoulders looking straight, head facing forward and eyes looking into the camera looking away, head turned to three-quarters, regardless of eye direction Sidenotes: # Seems to interact well with other loras # A good resolution to render at is /size:640x640 & /size:640x832 # these emotions can be combined: for example "gnashing grimacing screaming mouth" OR "squinting glaring eyes" # Mix and match for sometimes stunning, sometimes hilarious results. # Going overboards on weighting the emotions can lead to some serious body horror # You can also mix things up by including the emotions in your negative prompting Basic Prompt: /render portrait photo of 1girl, adorable face, pleasurable eyes, smiling mouth [blurry, lowres, worst quality] <realvis20>
inversion	enid-sinclair		female characters realistic semi-real	No
style	epic		semi-real .base models	No
style	epic-realism3		realistic .base models	No
inpainting	epic-realism3-inpainting	No		No
lora	eudoll		characters female anatomy realistic white	A non-specific character model of a sunny beautiful European woman' portrait, trained on 4,500 images. Can create a wide variety of characters, in many ethnicities and hair colors. It is recommended with <chilloutmix> though may bias towards Asian, so put that in your negative prompt if you're going for a different look
lora	evangeline		characters realistic female white	No
sci-fi	evangelion		mecha anime armor	A mecha LoRA model inspired by Evangelion, by LeoSam. Check the triggers to activate the different robot models. Notes from the creator: This is a LoRA that can generate EVA-like general-purpose humanoid decisive weapons. The tags EVA 00, EVA 01, EVA 02, and EVA 08 are respectively associated with machine number zero, machine number one, machine number two, and machine number eight. The No. 1 and No. 2 machines are more effective because they have more training images. The second is the zero machine, and the generation effect of the eight machine is average. In addition, I tagged all the training set pictures with the EVAGOD tag, so when this tag is called, a fusion-style humanoid weapon can be generated. Please note that because of the various sources of training photos in the model, it cannot accurately reproduce the details of the body in the original anime, but it has a significant EVA style in style
style	expressionism		semi-real .base models paintings art style	No
lora	eyeroll		anatomy expressions	No
anime	ez-viera		furry anime illustrations hentai gaming	EZ-Viera by iDeNoh is a LoRA trained on images of Fran from FF12, which ended up as a general model for creating Viera (meaning rabbitfolk). The creator recommends it on any kind of base model, from real to fantasy. At a low weight, you can get a general bunny woman. At the higher end weights, you definitely get Fran
lora	face-orgasm		expressions	No
effect	face-poser		expressions porn oral	No
pose	facesitting-pov		poses porn	This Lora is made to create an "imminent facesitting" low-angle shot. At weaker strengths, and in between ideal outputs, it will generate other nice low angles. Sweet spot is around 0.7-0.8, consistency may be improved by the "from below" tag. Trained on NAI from 91 custom images, generated from a small number of photo sources. Most of the example photos were anime but some semi-real as well. Known Issues: Often causes distortion of the nose. Inpainting works as a fix.
lora	fakemon		gaming pokemon nintendo	No
lora	fakemon2		gaming nintendo pokemon	No
lora	fakemon3		pokemon illustrations anime retro gaming	No
lora	fanart		cartoons illustrations	No
clothing	fantasy-armors		clothes fantasy armor	No
sci-fi	fantchar		characters cyberpunk 3d armor creatures fantasy monster girls realistic sci-fi semi-real	Fantastic Characters V4 by moesah Introducing Fantastic Characters! There' witches, and robots, and creatures, oh my! (and much, much, more) Need to make a new character? Maybe it needs to be something from every day life, like a construction worker. Or maybe you need something more fancy like a futuristic cyberpunk ninja. How about a nurse with robotic implants? Aliens, humanoid creatures, robots, mobsters, cowboys... the list goes on and on. Mix and match. Impress your friends and relatives! Make a new Fantastic Character today! Use the fantchar keyword to trigger this model. The trigger word is not strictly necessary, but it does seem to help with the finer details. This model is designed for showing characters in portraits, solo images, posing for the camera, etc. The model is tuned primarily towards realism and 3d, but it can generate some more cartoony or artsy stuff sometimes. You'll get the best results from characters with humanoid anatomy (two arms, two legs, torso, head, etc.), though it' good at making lots of different types of creatures and characters in that general shape (see the example images.) I prefer to use this model with Realistic Vision V4.0 as the base for a more realistic look, and A-Zovya RPG Artist Tools (V3+VAE) as the base for a more fantasy look. I've also seen it spit out nice results with others like DreamShaper 7, Deliberate V2, and RevAnimated. Here' the info on the latest version: V4: trained with about 6x the images as V1. Can make the widest variety of characters. The overcooked look from V3 appears to be fixed. Increased the dimension to 128 for this version since I ended up doubling the number of source images. Final notes and tips: I used Clip Skip 1. Anime base models might be less fantastic, but I haven't tried it so let me know. There' very little "Action" going on in the source images, so you might want to avoid using verbs in your prompt. ("Viking with axe" is better than "Viking swinging axe", for example). Be careful with your negative prompts, especially if you want something more on the 'creature' side than the 'human' side. There are several common negative prompts that will conflict or work against this Lora, so add them in as needed and remember that less is sometimes more. Weights between 0.8 and 1.0 seem to work best. Most of the source images are in the 768x512 or 512x768 buckets, but there' lots of buckets and most sizes seem to work fine for me. I didn't use any explicitly NSFW images in the dataset, so if you're getting lots of topless women, that probably isn't coming from this model. There are some attractive cosplayers in the dataset though, so it might be biased more towards making hot bartenders than plump barmaids. I did try to include a diverse mix though, so it might be fine. Your mileage may vary.
lora	farsider		cartoons illustrations memes	No
clothing	fashiongirl		clothes female anatomy anime semi-real	Fashion Girl by ZakP (creator of YesMix) is a general beauty enhancer for portrait and body shots for women, with optional triggers for age and lipstick. Use it to make faces more beautiful. It was trained on 100 photos of fashionable women. Despite the name, it can also enhance nudes and general anatomy. The latest version skews more towards anime, however is suitable for semi-real as well. Use base models like <aohard> and <hassaku> and <counterfeit>. While it is possible to use it in realistic renders, you'll want to drop the strength by a lot
inversion	fast-negative	No		Fast Negative Embedding Version 2 by Lykon is a merge of common embeddings, including Nerf' Negative Hands. Works well as an all-purpose quality booster. The trigger MUST be included within negative brackets with a weight of negative 1 to two. Example: [[<fast-negative:-1>]]
anime	fate-astolfo		characters anime asian fate male	A LoRA of Astolfo (from the Fate series) by AA, this is version 4. Use it with any anime base model. Seems to respond to nudity well While it does a good job of generating Astolfo without any physical descriptor tags, adding some can help. For instance, I've found that it struggles with doing his eye color naturally, so adding 'purple eyes' improves it
anime	fate-rin		anime characters female fate	LoRa dedicated to Tohsaka Rin from Fate, one of the classic tsundere queens. 1) Usage: # Weight: 0.1-0.4 # Use: “/more /strength:0.4 /size:1024x1024 /images:5 /guidance:15 <your model>” on the picture you like to improve quality. 2) Recommended base models: # <av5> the most canon look. Example: /render /guidance:6 /sampler:dpm2m /size:768x832 1girl, tohsaka rin, solo, long hair, thighhighs, skirt, blue eyes, black thighhighs, black hair, ((masterpiece, good quality, intricate details)), [lowres, blurry, worst quality] <fate-rin:0.35> <av5> # <revani> good results. Example: /render /guidance:9 /sampler:k_euler_a /size:768x832 1girl, tohsaka rin, solo, long hair, thighhighs, skirt, blue eyes, black thighhighs, black hair, masterpiece, good quality, [lowres, blurry, worst quality] <fate-rin:0.3> <revani> # <counterfeit> good results. Example: /render /guidance:8.5 /sampler:k_euler_a /size:768x896 1girl, solo, long hair, skirt, pantyhose, tohsaka rin, shirt, black skirt, white shirt, homurahara academy school uniform, open clothes, ribbon, coat, blue eyes, red coat, brown vest, long sleeves, collared shirt, black pantyhose, school uniform, bangs, vest, brown hair, open coat, ((masterpiece, good quality, simple background)), [lowres, blurry, worst quality] <fate-rin:0.2> <counterfeit> 3) Recommended prompts to obtain different looks: # Normal Rin: 1girl, tohsaka rin, solo, long hair, thighhighs, skirt, blue eyes, black thighhighs, black hair # School uniform Rin: 1girl, solo, long hair, skirt, pantyhose, tohsaka rin, shirt, black skirt, white shirt, homurahara academy school uniform, open clothes, ribbon, coat, blue eyes, red coat, brown vest, long sleeves, collared shirt, black pantyhose, school uniform, bangs, vest, brown hair, open coat
anime	fate-saber		anime characters female	No
clothing	fbi-uniform		clothes memes	No
inversion	fcneg			No
lora	fear-expression		expressions	No
pose	feetpose		poses feet fetish	A general feet pose in different angles. A weight between 0.6 to 0.8 with an anime model like <av45> works best. The creator, Notaledeji, recommends the following prompt structures: 1girl, sitting, feet up, soles, YOUR DETAILS, <lora> <base> Example 2: 1girl, sitting, feet up, soles, brown hair, blue eyes, white bikini, bedroom, <lora> <base>
pose	feetpose2		anatomy poses feet fetish	A seated woman with her feet towards the camera, showing off her soles. Recommended weight is 0.6 with URPM as the base
anime	felicia-darkstalkers		gaming characters female monster girls anime	No
lora	female-masturbation	No		No
pose	female-pov		poses porn female	No
anime	femboi		anatomy trans anime male penis porn realistic lgbtq	A Femboy 'Concept' LoRA. Turns the subject(s) of a generated image into an erect femboy/crossdressing femboy virtually without fail, and creates a strong lean towards: Visible assholes/gaping assholes 'before' and 'after' sex. Lean-yet-toned body/androgynous body aesthetic. Anal sex with an off-screen partner. Group sex with the femboy subject(s) as the focus. Having the subject(s) wear heels, toeless legwear, chokers, and ball bras. Demure posturing such as a hand on one' chest, and presenting with spread legs. It cannot do flaccid penises or chastity/it has very little training with them. Highly recommend that you do gens in dimensions between 512x512~640x640, and then use hires fix to upscale+re-gen them, as it has far more success making coherent images within these ranges compared to 768 and beyond. This lora should function well at 1.0 strength, especially in simpler prompts, but you can try adjusting it 0.6~1.2 for effect, as well as tweaking the keyword directly (such as femboi:1.1). Adding 'bent over' or 'from behind' as 2nd in the prompt with 'femboi' as 1st should consistently generate ass shots/'cocktails'. Use 'open-toe high heels' or 'wedge sandals/sandals' rather than high heels, as the lora has been trained to work with open-toe footwear specifically. Adding 'pasm' near the end of a prompt will cause those 'currently orgasming' lines to appear around the subject(s). Adding 'heart eyes' will cause accurate heart icons to form around the subject(s), and when used along with 'pasm' they will further increase the chance and accuracy of an anal sex image. 'heart' as a word alone in the prompt will just add the accurate heart icons. Adding 'logo:2' as a negative to a prompt can augment images. I say augment because it isnt guaranteed to enhance quality and sometimes just changes the image, but it can and may be more effective on images that seem weaker than average with art quality or composition. This lora was trained on CounterfeitV2.5, but can likely offer solid performance on other anime-style models with 50 steps and /guidance:9
anime	femboi2		anatomy trans anime male penis porn realistic lgbtq	Otoko No Ko LORA, from behind, facing viewer, on stomach and missionary sex. Use at your own risk, requires some gacha. Use at strength 0.6-1.0. Use activator 'Otono Ko No'. With a more focused dataset and manual tagging, this could be a lot better, so I either expect better LORAs to be coming out for this concept in the near future, or a model and sampler more forgiving of the concept. The issue with this concept is that AOM2 and SDE Karras aren't as friendly to penises on a solo character as base SD/Euler. Overall it works, but with a good amount of gacha. If your generation only contains testicles from behind, bump LORA strength up slightly. Requires some seed gacha, so hit the seed recycle button and continue to gen
effect	femjac		effects porn	Strengthen the performance effect, and it is recommended to use it with other posture LORA.It performs better on the positive side than other perspectives. The liquid position may be strange from other perspectives. A weight of 1 is fine with most anime base models at a guidance of 7
fantasy	ff-tifa		characters female gaming final fantasy	Tifa Lockhart <All Outfits> LoRA v.offset by Lykon Author' comment: "At some point I noticed nobody had yet made a LoRA of her because there are already so many decent TIs, so I decided to make one, and being this a LoRA, I could fit ALL Tifa' outfits. This one triggers with tifa lockhart but can also accept game/movie names to specify which Tifa (e.g: final fantasy vii advent children for her AC outfit + descriptive tags) and dress tags for the official Remake outfits (e.g: tifa lockhart' refined dress). Check pics on Danbooru for all the tags, or the examples I provided. It also gets very easily her cowgirl outfit and, with a bit of failed attempts, her green dress from the flashback (kind of). Of course it can do her OG FF7 outfit."
lora	ff7		final fantasy gaming	No
pose	finger-in-mouth		poses porn oral	No
pose	fingering		poses porn	A pose lora for female masturbation, has triggers for an optional self boob grab. The creator notes that it works best at two resolutions: 512x768 or 512x896. To make the clothes go away, add nsfw, nude prompts, etc. Also works well with 1girl, pussy, juice, cum, blush
pose	fingering2		poses porn	No
anime	fist-of-the-north-star		anime retro	Fist of the North Star / Haruto No Ken LoRA by AI Fantasy Art. Notes from the creator: This LoRA has been trained to emulate the style of the popular 80' anime Fist Of The North Star aka Hokuto No Ken or ???? in Japan. pair with: <deliberate2> recommended settings: /sampler:k_euler_a /steps:60 /size:512x768 /guidance:3 for better quality: /highdef /facelift /anime
pose	fitroom-selfie		female poses environments	a Lora that produces photos of females taking selfies in fitting rooms Recommended settings for getting good renders: # Guidance works best at aprrox. 8 # Weight is best between 0.8 and 1 # Works with most realistic based models also works with un-realistic models Sidenotes: # Seems to interact well with other loras including character loras (works great with <flashingtits> and <skirtlift> # A good resolution to render at is /size:512x768 # This lora does the pose and bodies well but can struggle with faces...the more descriptive you are, the better the result Basic Prompt: /render 1girl, fitting room, selfie, <fitroom-selfie> <edge>
lora	flashing-tits-aiguru	No		No
pose	flashing-tits-bikini		poses boobs porn	Bikini Pull by Waifu Wizard Wong is a flashing boobs bikini pose for realistic photos, where the bikini top is still on but pulled aside to reveal the nipples. Works with both anime and realistic models at a weight of 0.4
pose	flashingtits		poses porn memes boobs	Shirtlift is a <liberty> based LoRA for realistic images sourced from the internet. It works with anything as a base model and should be pushed above a weight of 1 for stubborn situations, like statues and astronaut uniforms. You know, for science.
effect	flashingtitsframes		effects porn boobs	Flashing Tits Frames is a LoRA that creates single (non-animated) frames that can be transferred into GIF making software. The model creator aDDont says "the results are always unstable, but flexible". Check the triggers for various options, like being able to render 2x2 frames. A weight of 0.7 should suffice and works well with semi-real models like Dreamshaper and Deliberate, try guidance at 7 if results look too crunchy
effect	flashlight		effects lighting	This LoRA is the converted version of an idea to make images not "beautiful", but ... different. It is only used to make people or things visible in a dark environment. The correct application for it is therefore clearly in the images of: night situations (streets, alleys, parks, ...) dark rooms (discos, clubs, cellars, apartments without electricity, ...) generally dark places (forest, caves, castles? ... ) Try it with a realistic photo base
effect	floral-patterns		graphic design effects	Floral Patterns LoRA by alc15492 creates abstract trees, leaves, flowers, and butterfly style patterns when used with a base model like <anylora> or similar. Try it at a weight below 0.8, and keep lowering it if it looks "overcooked". At a very low weight it can also be used to spice up a portrait or environment as an effect
sci-fi	flying-cars		sci-fi environments objects vehicles	Amazing Landspeeders & Futuristic flying cars v1.0 by fitCorder # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low # Mangled limbs, general chaos ? Some LoRas can’t handle rendering with resolutions bigger than 512x512 // Also make sure your prompt isn’t too complicated for this LoRa # No visible effect ? Check if the LoRa has any triggers that need to be added in order to activate it # Good full models to try out LoRas with: anime ? <anylora> <counterfeit> <breakanime> comics ? <comics-vision> <icomix> <misstoon> semi-real ? <revani> <lyriel16> <dreamshaper6> realistic ? <majic> <bra5> <chilloutmix> <edge>
pose	fondle		poses porn hentai	No
object	food-photos		realistic objects	No
pose	footlick		fetish anime feet poses	Use the following prompt to activate this LoRa properly: 1girl, close up, tongue out, saliva, soles, YOUR DETAILS, <lora> Example : /render /size:768x768 ((masterpiece, good quality, detailed face)) 1girl, close up, tongue out, saliva, soles, nurse, red eyes, hospital, <footlick:0.9> [[[lowres, blurry, worst quality]]] <av5> It was trained at 768x768 resolution and works much better if you generate at 768x768. The pose is very specific, so there is little room for variety.
lora	fps		retro guns gaming	This lora, created by a community member, was designed to insert firearms into the foreground to make any image appear as if it is from a FPS game or some kind of body cam footage. Sometimes guns may not appear, try adjusting lora weight and perhaps the base model used if this happens.
pose	frame-binder		bondage poses	No
anime	frontline-a545		characters gaming girls-frontline female anime	Girls' Frontline-A545 v1.0 by L A X # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low # Mangled limbs, general chaos ? Some LoRas can’t handle rendering with resolutions bigger than 512x512 // Also make sure your prompt isn’t too complicated for this LoRa # No visible effect ? Check if the LoRa has any triggers that need to be added in order to activate it # Good full models to try out LoRas with: anime ? <anylora> <counterfeit> <breakanime> comics ? <comics-vision> <icomix> <misstoon> semi-real ? <revani> <lyriel16> <dreamshaper6> realistic ? <majic> <bra5> <chilloutmix> <edge>
anime	frontline-ar57		characters gaming girls-frontline female anime	Girls' Frontline-AR57 v1.0 by L A X # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low # Mangled limbs, general chaos ? Some LoRas can’t handle rendering with resolutions bigger than 512x512 // Also make sure your prompt isn’t too complicated for this LoRa # No visible effect ? Check if the LoRa has any triggers that need to be added in order to activate it # Good full models to try out LoRas with: anime ? <anylora> <counterfeit> <breakanime> comics ? <comics-vision> <icomix> <misstoon> semi-real ? <revani> <lyriel16> <dreamshaper6> realistic ? <majic> <bra5> <chilloutmix> <edge>
anime	frontline-fp6		characters gaming girls-frontline female anime	Girls' Frontline-FP-6 v1.0 by L A X # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low # Mangled limbs, general chaos ? Some LoRas can’t handle rendering with resolutions bigger than 512x512 // Also make sure your prompt isn’t too complicated for this LoRa # No visible effect ? Check if the LoRa has any triggers that need to be added in order to activate it # Good full models to try out LoRas with: anime ? <anylora> <counterfeit> <breakanime> comics ? <comics-vision> <icomix> <misstoon> semi-real ? <revani> <lyriel16> <dreamshaper6> realistic ? <majic> <bra5> <chilloutmix> <edge>
anime	frontline-honey		characters gaming girls-frontline female anime	Girls' Frontline-Honey Badger v1.0 by L A X # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low # Mangled limbs, general chaos ? Some LoRas can’t handle rendering with resolutions bigger than 512x512 // Also make sure your prompt isn’t too complicated for this LoRa # No visible effect ? Check if the LoRa has any triggers that need to be added in order to activate it # Good full models to try out LoRas with: anime ? <anylora> <counterfeit> <breakanime> comics ? <comics-vision> <icomix> <misstoon> semi-real ? <revani> <lyriel16> <dreamshaper6> realistic ? <majic> <bra5> <chilloutmix> <edge>
anime	frontline-ltlx7000		characters gaming girls-frontline female anime	Girls' Frontline-LTLX7000 v1.0 by L A X # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low # Mangled limbs, general chaos ? Some LoRas can’t handle rendering with resolutions bigger than 512x512 // Also make sure your prompt isn’t too complicated for this LoRa # No visible effect ? Check if the LoRa has any triggers that need to be added in order to activate it # Good full models to try out LoRas with: anime ? <anylora> <counterfeit> <breakanime> comics ? <comics-vision> <icomix> <misstoon> semi-real ? <revani> <lyriel16> <dreamshaper6> realistic ? <majic> <bra5> <chilloutmix> <edge>
anime	frontline-vhs		characters gaming girls-frontline female anime	Girls' Frontline-VHS -Keeper v2.0 by L A X # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low # Mangled limbs, general chaos ? Some LoRas can’t handle rendering with resolutions bigger than 512x512 // Also make sure your prompt isn’t too complicated for this LoRa # No visible effect ? Check if the LoRa has any triggers that need to be added in order to activate it # Good full models to try out LoRas with: anime ? <anylora> <counterfeit> <breakanime> comics ? <comics-vision> <icomix> <misstoon> semi-real ? <revani> <lyriel16> <dreamshaper6> realistic ? <majic> <bra5> <chilloutmix> <edge>
sci-fi	frost-punk		effects sci-fi fantasy armor	No
lora	frozen-elsa		characters female anime cartoons	No
pose	fucked-silly		hentai poses	Simple LoRa that’s good at its job. It contains one pose and will stick to it. 1# Works best with weight around 0.8-1. 2# Tested on anime base models like <aom3>, <aohard>, <countefeit>. 3# Best rendering resolution is around 640x640. 4# Is surprisingly easy to use with different characters LoRas. 5# Needs quite a lot of negative prompts to produce quality results.
anime	fulldive-mizarisa		female characters anime	Mizarisa - Full Dive v1 by Zunopziz Recommended settings for getting good renders: # Weight ~0.2-0.35 # Keep rendering size small, for example 512x768 # Can be used with most anime base models like <counterfeit>, <anylora> # /parser:new /clipskip:2 (optional) Sidenotes: # Add prompts from the triggers accordingly
style	fumo-doll		art style anime toys	No
anime	furry-eevee		anime creatures furry pokemon nintendo	Lora dedicated to eevee and all its forms. The author suggests it works best with <yiffymix2>, but it can handle other anime models as well. For delicate touch use weight 0.1-0.3, and 0.5-0.8 for normal use. Putting the prompt ‘anthro’ in negative results with a more ‘feral’ look.
anime	furry-girl		characters furry female anime creatures	A non-specific female furry character in an anime style, as <aohard> was the base. A weight of 0.6 to 1 should be fine. Creator Kizarou does not recommend using it with Meinamix or Cetus, however
lora	futa		trans penis porn anatomy	No
lora	futa-sm		trans penis male porn realistic lgbtq	A lora that can be used on both males and females. Describing the scene in detail while using it produces better-quality results. Avoid using too many trigger words at once. Recommended weight 0.6-0.8. You can also specify between: "tiny penis" and "small penis" with either "flaccid" or "erection". "Small testicles", "medium testicles" or "large testicles". "Small breasts", "medium breasts", "big breasts", "huge breasts". Other prompts: "tucked penis", "penis in panties", "erection under clothes", "bulge", "panty bulge", "female pubic hair", "foreskin", "phimosis", "penis ribbon", "penis ring" (although these last tags 2 could be hard to use) and "penis out" (for penis poking out of panties, its not easy to use, sadly). "from behind" (remember to include "ass", "anus", "perineum" or "penis on panties" and "panty bulge" based on what you want), "from above" and "from below" should also work.
lora	futa-veiny		trans futa penis porn anatomy	No
lora	futa-xl		trans penis male porn realistic lgbtq	No
lora	futa2		trans penis porn lgbtq	No
lora	futa3		penis trans lgbtq porn hentai	No
lora	fuyumi-todoroki	No		No
effect	gachasplash		anime effects	GachaSplash is a LoRa that will give our renders an interesting background effect. Unfortunately, it’s not entirely compatible. It still works and can produce good-looking results, but despite following the template created by the author, obtaining the desired background effect seems random. Template: [(white background:1.5)::5], isometric OR hexagon, 1 girl, mid shot, full body, <add your background prompts here> Try it out with <counterfeit>,<counter25>, or <av5> models. Each of them produces amazing results. Add wood and stone as negative prompts if you see too many of those elements. This can also be applied to certain clothing elements if the LoRa keeps producing characters with them. Suggested resolution: 640X640, then use /highdef on the best picture. In this case, treat the triggers more like suggestions of prompts to try out.
lora	gadot		characters female realistic white	No
style	galaxytime		semi-real .base models fantasy environments architecture interiors	No
pose	gape		anatomy poses porn butts	No
clothing	garter-belts		clothes	To get the best of this LoRa you will need more detailed prompts describing the clothing. For example: highly detailed face and eyes, photorealistic, smiling, garterbelt, a woman in a lingerie posing for a picture, wearing the garterbelt, colorful, floral embroidery, frilled skirt, black garterbelt with red embroidery and multiple leather straps, multiple iron chains attached to rings
clothing	garter-belts2		clothes porn	This is V2 aka "Super Sexy" version of Garter Belts by EDG, which produces more realistic results. Due to the complexity of the garment you might want to try these templates: Masterpiece, absurdres, fine detail, HDR, highly detailed face and eyes, photorealistic, smiling, <lora:garter-belt2>, garterbelt, black fabric, fabric and metal straps, leather straps, embroidery, woman wearing the garterbelt, lower body focus, front picture <chilloutmix> or change it up a bit: Masterpiece, absurdres, fine detail, HDR, highly detailed face and eyes, photorealistic, smiling, Woman wearing a black garterbelt with red embroidery and multiple leather straps,back view, lower body focus, multiple iron chains attached to rings
lora	gelato-style		effects art styles	No
lora	gelatostyle	No		No
style	gem-punk		effects art style	This lora is designed to add beautiful gemstones to whatever is used as a prompt. A castle encrusted with emeralds, a dragon with ruby scales the world is your shiny bejeweled oyster. Lora works great with revani at default weight.
inversion	gender-demasculate		anatomy lgbtq trans	Gender Slider: DeMasculate by Zoyva is a Textual Inversion that is supposed to "Place in your negative prompt at strength of 1.0 to 1.3. This will removes beards, penises and other masculine features. Great for removing males from a scene." In our testing, the effect is very mild, so depending on the model and character you may only notice some slight changes and not a full-on gender swap
inversion	gender-slider	No		No
anime	genshin-candace		genshin impact gaming female anime characters hentai	The hydro anime female gaming character from Genshin Impact. As it was trained on AOHard, this model is intended for hentai
anime	genshin-dehya		female characters genshin impact anime	No
anime	genshin-eula		female characters genshin impact anime gaming	No
anime	genshin-ganyu		genshin impact characters female gaming anime	No
anime	genshin-hutao		genshin impact characters female gaming anime	No
anime	genshin-kokomi		characters anime female gaming genshin impact	No
anime	genshin-kuki		genshin impact characters female gaming anime	No
lora	genshin-mona		characters female gaming genshin impact	No
anime	genshin-nilou		genshin impact characters female gaming anime	No
anime	genshin-noelle		characters female gaming anime realistic genshin impact	No
anime	genshin-raiden		female characters genshin impact anime	No
anime	genshin-shenhe		female characters genshin impact anime gaming	No
lora	genshin-sucrose		characters gaming female genshin impact asian	No
anime	genshin-yelan		female characters genshin impact anime gaming	No
anime	ggear-ino		characters female gaming anime	I-No (Guilty Gear) v1.0 by Murky # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low # Mangled limbs, general chaos ? Some LoRas can’t handle rendering with resolutions bigger than 512x512 // Also make sure your prompt isn’t too complicated for this LoRa # No visible effect ? Check if the LoRa has any triggers that need to be added in order to activate it # Good full models to try out LoRas with: anime ? <anylora> <counterfeit> <breakanime> comics ? <comics-vision> <icomix> <misstoon> semi-real ? <revani> <lyriel16> <dreamshaper6> realistic ? <majic> <bra5> <chilloutmix> <edge>
anime	ggear-jacko		characters gaming female anime	Jack-O' Valentine (Guilty Gear) v1.0 by Murky # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low # Mangled limbs, general chaos ? Some LoRas can’t handle rendering with resolutions bigger than 512x512 // Also make sure your prompt isn’t too complicated for this LoRa # No visible effect ? Check if the LoRa has any triggers that need to be added in order to activate it # Good full models to try out LoRas with: anime ? <anylora> <counterfeit> <breakanime> comics ? <comics-vision> <icomix> <misstoon> semi-real ? <revani> <lyriel16> <dreamshaper6> realistic ? <majic> <bra5> <chilloutmix> <edge>
anime	ggear-may		characters gaming female anime	May (Guilty Gear) v1.0 by Murky # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low # Mangled limbs, general chaos ? Some LoRas can’t handle rendering with resolutions bigger than 512x512 // Also make sure your prompt isn’t too complicated for this LoRa # No visible effect ? Check if the LoRa has any triggers that need to be added in order to activate it # Good full models to try out LoRas with: anime ? <anylora> <counterfeit> <breakanime> comics ? <comics-vision> <icomix> <misstoon> semi-real ? <revani> <lyriel16> <dreamshaper6> realistic ? <majic> <bra5> <chilloutmix> <edge>
style	ghibli		anime .base models art style art style	No
style	ghostmix		.base models anime mecha sci-fi	GhostMix Version 2 is a full checkpoint base model focused on intricate artworks, inspired in part by the biomechanical style of Ghost in the Shell anime. It can also produce interesting effects with fractal and watercolor tags. Like many anime base models, it uses modified Danbooru tag method. The main deviation is a signature character that can be invoked with "1mechanical girl". This character appears more often when the aspect ratio is 512x768 or similar, and a guidance of 7 or less is recommended overall.
sci-fi	ghoul		post-apocalyptic sci-fi creatures gaming	Designed by Community Member Eggs-Benadryl. This lora is designed to generate sentient ghouls from the Fallout series, a popular video game franchise. It can create both animated and realistic models of these irradiated humanoids, with various levels of decay and mutation. You can adjust the weight parameter from 1 to 1.5 to control the effects. To avoid unwanted features such as glowing eyes or zombie-like effects, you can use the negative prompt option to exclude them from the generation process. This lora is ideal for fans of the Fallout series who want to create their own ghoul characters or scenarios.
lora	gigachad		characters memes male white	The chaddest of them all - GIGACHAD Lora strength closer to 1 will create the ultimate gigachad. For more flexibility, consider lowering the value. In case the renders are filled with artifacts, try changing the base model or lowering guidance. It can be used both for realistic, and anime renders.
sci-fi	giger		semi-real art style horror sci-fi anatomy creatures effects fantasy	No
lora	gigi-rivera	No		No
lora	gina-carano		characters female realistic white	No
pose	glory-hole1		poses porn environments	No
pose	glory-hole2		poses porn environments	No
pose	gloryhole-close		porn poses oral	a lora that creates renders of women giving gloryhole oral sex Recommended settings for getting good renders: # Guidance at approx 7.0+ # Best weight right around 0.3 # Works great with realistic and non-realistic models # requires relatively complex prompting Sidenotes: # Seems to interact well with other loras # Render at low sizes and use /more /size /strength to upscale # Multiple penis can be solved by using "penis"as a negative prompt # adjusting the weights of "wall" and "penis coming out of the hole in the wall" will allow more control of this lora Example Prompt: /render GloryHoleCloseQuiron pose, (look at viewer), 1girl, penis, solo, fellatio, sexually suggestive, close up, (wall:1.8), (penis coming out of the hole in the wall:1.6), <gloryhole-close:0.3> [ blurry, lowres, worst quality] <urpm13>
effect	glow-mecha		mecha anime female lighting effects armor	This LoRa has two modes. One makes cute mecha girls; the other adds stunning glowing details. In both cases, it’s very good at making nice, long legs. How to use: # Keep the weight low (0.2-0.4); the quality deteriorates very fast above 0.5 # Use “mecha” in prompts to activate LoRa’s mecha-making properties. This works great with anime models such as <cetus>, <counterfeit>, etc. # Use “glowing” & “lighting particles” to add awesome sci-fi lighting effects to your characters. Tested with <revani>, <lyriel> and <aio>. # All triggers can be used together. # “spread legs” is not a typo; this is the prompt taken from the author’s examples.
effect	glowing-runes		objects lighting effects	Adding glowing runes makes everything more awesome, and this LoRa proves it. Easy to use, works with most base models, but looks best with semi-real ones like <revani>. You can control the number of runes/glowing with weight. In order to make the runes appear, you must include one of the triggers. Red and Blue variants are certain, while other colors depend on various factors.
anime	goblin-girls		characters female monster girls anime fantasy	No
fantasy	goblins		creatures characters male fantasy memes anime hentai	No
lora	god-neko		characters female asian realistic boobs	No
anime	goku		characters anime male dragonball	No
style	gopher-alpha		art style anime semi-real	GopherAlpha is a variant of a set of LoRAs created by the model creator known as GopherReflection on Civitai and Reddit. Focused on female portraits, the style adds colorized comic-style inks, paintings, or vivid colors with very curvy proportions. Our testers found that strength is inconsequential, however it may react differently when other LoRAs are also used. There are other variants of Gopher, including a grinny Meme style, a more semi-real version called Theta, but they don't look radically different from each other.
style	gopher-beta		art style anime semi-real	GopherBeta is a variant of a set of LoRAs created by the model creator known as GopherReflection on Civitai and Reddit. Focused on female portraits, the style adds colorized anime-styled inks, paintings, or vivid colors with very curvy proportions. Use with weight between 0.5-0.85. There are other variants of Gopher, including a comic-like Alpha style, a more semi-real version called Theta, but they don't look radically different from each other.
clothing	gothgirl		characters female porn clothes realistic semi-real anime	Goth Girl is a non-specific black haired black clothing character, usable in real, semi-real, or anime base models. There are other gothgirl loras, and some are tamer or focused on frilly clothes. This one is the porn one, specifically designed for skimpy leather outfits, choke collar, and nudity. The creator, Schneed, recommends a strength from 0.6 to 0.8 with a base
clothing	gothgirl2		characters female clothes realistic semi-real anime	Officially known as Goth Gals, this is a fashion Lora based on Harajuku style black maid costumes, long black hair, black makeup, garters, stockings, that sort of thing. Works well with anime, semi-real, and real photographs. Use our other, very different nsfw goth fashion LoRA specifically for the kinkier goth stuff.
clothing	gothgirl3		characters female clothes	No
style	gothic-aesthetics		effects art style architecture environments horror	No
clothing	gothic-fashion		clothes	Good for adding details to gothic-styled dresses and attire. By itself, it may not do much, but after describing the clothes in more detail, it will help to bring out their more intricate side. 1# Works best with weight around 0.5-0.9. The more details you need, the higher you should set it. 2# Works great with anime and semi-real base models. 3# Can handle rendering in bigger sizes. Example: /render /size:832x1024 1girl, absurdly long hair, air bubble, breasts, bubble, gothic-lolita dress, eyebrows visible through hair, floating hair, full body, thighhighs, hair between eyes, hairband, long hair, long sleeves, looking at viewer, medium breasts, red dress, eyes, ripples, silver hair, sleeves past wrists, solo, standing, underwater, very long hair, water, waves, (((masterpiece, realistic, good quality, absurdres, intricate, ethereal, detailed, breathtaking, lovely))), [[lowres, worst quality, blurry, extra digits, low quality, poorly drawn, sketch, soft lines]] <gothic-fashion:0.9> <counterfeit>
style	gothic-punk		effects art style	This lora is designed to give a gothic look to anything and everything. It adds gothic architecture, clothing and vibes to anything it touches. Works well at default weight. Try with any model. Revani works especially well
anime	grabby-hands		porn anime hentai	No
anime	granblue-hecate		characters anime female	Hecate (Granblue Fantasy) v 0.5.1 (collab) by deatharms2010 There are three ways to use this lora: 1 - use the prompt “hecate” 2 - use those prompts: “orange eyes, blonde hair, fox ears, cat eyes, black dress, ahoge, ornaments, long hair” 3 - use it all together “hecate, orange eyes, blonde hair, fox ears, cat eyes, black dress, ahoge, ornaments, long hair” General lora guidelines: # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low # Mangled limbs, general chaos ? Some LoRas can’t handle rendering with resolutions bigger than 512x512 // Also make sure your prompt isn’t too complicated for this LoRa # No visible effect ? Check if the LoRa has any triggers that need to be added in order to activate it # Good full models to try out LoRas with: anime ? <anylora> <counterfeit> <breakanime> comics ? <comics-vision> <icomix> <misstoon> semi-real ? <revani> <lyriel16> <dreamshaper6> realistic ? <majic> <bra5> <chilloutmix> <edge> "
lora	gravity-falls		cartoons comics illustrations	No
lora	greatscott		characters male	No
lora	green-heart-neptunia	No		No
lora	greg-abbot		male characters memes	No
lora	greta-thunberg		characters female memes realistic	Greta Thunberg likeness, created by model maker GRTmate who says "all images trained are 18 and over, face and upper body" for all your activist meme needs
style	gta5		gaming .base models cartoons	No
lora	gulnazar		characters female asian chinese	Gulnazar is a Chinese actress & model, born on 2 May 1992 in Ürümqi, Xinjiang, China. She is particularly known for her role as 'Xiao Wei' in the Film “Police Story 2013"
object	gun-ak47		objects guns	A working gun model of the AK47, however seems hard to control based on internet comments. The creator recommends the base model as <chilloutmix> as the main go-to
object	gun-blyat47		objects guns	No
object	gun-glock17		objects guns	No
object	gun-hk416		objects guns	No
object	gun-m4a1		objects guns	No
object	gun-m4carbine		objects guns	No
pose	gun-point		poses guns anime comics	No
object	gun-shotgun		objects guns	No
object	gun-shotgunpump		objects guns	No
object	gun-sniper82		objects guns	No
sci-fi	gundam-fleet		sci-fi environments objects spaceships	gundam/EFSF fleet type v1.0 by 494369066868 # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low # Mangled limbs, general chaos ? Some LoRas can’t handle rendering with resolutions bigger than 512x512 // Also make sure your prompt isn’t too complicated for this LoRa # No visible effect ? Check if the LoRa has any triggers that need to be added in order to activate it # Good full models to try out LoRas with: anime ? <anylora> <counterfeit> <breakanime> comics ? <comics-vision> <icomix> <misstoon> semi-real ? <revani> <lyriel16> <dreamshaper6> realistic ? <majic> <bra5> <chilloutmix> <edge>
style	guofeng		semi-real .base models fantasy environments chinese asian	GuoFeng is trained on gorgeous antique Chinese art, but in high resolution so looks like 2.5D gaming arts. Makes near-realistic portraits and landscapes, very saturated and soft images. Plays well as a base for other art style loras
anime	gurran-yoko		female anime characters	# Outfit1 ponytail, scarf, bikini top only, short shorts, midriff, belt, fingerless gloves, skull hair ornament, elbow gloves, pink thighhighs # Outfit2 ponytail, underboob, cape, bandeau, hair ornament, strapless, tube top, midriff, white gloves, star (symbol), thighhighs
lora	gwen-stacy-sidecut		characters female comics cartoons	No
lora	gwyneth-paltrow	No		No
clothing	gym-outfit		clothes	Female Gym Outfit by Antonio Riolo2610. Recommended settings: Lora Weight 0.3 - 0.7 You can prompt something like "wearing red outfit" in addition to the optional triggers. Works with realistic base models like <chilloutmix> and anime base models, too.
style	gyokai		anime art style	Lora based on the art of Gyokai Ononoimoko (Dohna Dohna' illustrator). It has a prominent colorful effect and may be hard to control. Seems to work best when mixed at low weight with other character loras. Reacts strongly to NSFW prompts.
style	gyokai2		semi-real anime art style	No
lora	hair-afro		hair	No
effect	hair-braid		anatomy effects female hair	No
effect	hair-buns		anatomy effects female	A LoRA hairstyle that' somewhere inbetween Bjork' buns and Princess Leia' Space Buns, it' hair balled up on opposite sides like two rolled-up long pig tails. Try it with a base like Chilloutmix or similar, at a weight between 0.5-0.7
lora	hair-censorship	No		No
pose	hair-grabbing		poses porn hair	No
lora	hair-length-slider		hair	No
effect	hair-reflection		effects anime lighting hair	No
effect	hair-wet		anatomy effects hair	No
pose	handjob		poses penis porn	This LoRa is very sensitive, it works only at minimal weight ~0.01-0.03
pose	handjob-cuddling		porn poses	a lora that creates renders of women giving a closeup cuddly handjob Recommended settings for getting good renders: # Guidance at approx 7.0+ # Best weight right around 1 # Works best with <hassaku>. Will also work with other non-realistic models but results will vary. Feel free to try it with realistic models but results can be VERY iffy. # Works great with basic and elaborate prompt Sidenotes: # Seems to interact well with other loras, especially character loras # Render at low sizes and use /more /size /strength to upscale Example Prompt: /render <handjob-cuddling> 1girl, cuddling handjob, 1boy, penis,[worst quality, blurry, lowres] <hassaku>
pose	handjob-nursing		porn poses	a lora that creates renders of women giving a handjob and having her tits sucked Recommended settings for getting good renders: # Guidance at approx 7.0+ # Best weight right around 1 # Works best with <hassaku>. Will also work with other non-realistic models but results will vary. Feel free to try it with realistic models but results can be VERY iffy. # Works great with basic and elaborate prompt Sidenotes: # Seems to interact well with other loras, especially character loras # Render at low sizes and use /more /size /strength to upscale Example Prompt: /render <handjob-nursing> 1girl, 1boy, nursing handjob, (breast sucking:1.0), nipples [worst quality, blurry, lowres] <hassaku>
lora	handjob-nursing2			No
pose	handjob-pov		poses porn penis hentai	How to use: # Try it out with weight 0.5-0.9 # It works best with anime based models like <aom3>.
pose	handjob-sitting		porn poses	a lora that creates renders of women giving a sitting handjob Recommended settings for getting good renders: # Guidance at approx 7.0+ # Best weight right around 1 # Works best with <hassaku>. Will also work with other non-realistic models but results will vary. Feel free to try it with realistic models but results can be VERY iffy. # Works great with basic and elaborate prompt Sidenotes: # Seems to interact well with other loras, especially character loras # Render at low sizes and use /more /size /strength to upscale # Be sure to use "futa" in your negative prompt if you want to avoid female penises Example Prompt: /render<handjob-sitting> 1girl, 1boy, sitting, handjob, couple [worst quality,((futa)), blurry, lowres] <hassaku>
clothing	hanfu		clothes chinese asian	Beautiful traditional Chinese fashions, with emphasis on the clothes. This is an all-female model with five different variants. Additionally, words like long shan, short shan, pleated skirt, and songmo can yield good results. Check the triggers for more info. The included token is HANFU
lora	hannah-owo		female characters	This lora specializes in creating images of social media influencer HannahOwo. It is intended to produce realistic and diverse portraits of her. The lora focuses on her visual appearance. Use it at a ridiculously low weight of 0.1 for best results.
style	hardblend		realistic .base models realistic porn	Hard Blend is a photo realistic base model trained for diverse portraits -- a versatile photorealistic NSFW capable model, blended in part with Hassans and Realvis
anime	harley		anime characters comics female	Needs around 0.03 weight! Makes "New style" Harley Quinn. Play around with the weight maybe 0.035 works best with anime style loras something like <av45>
lora	harley-2019-series	No		No
lora	harley-margot	No		No
style	hasdx		semi-real .base models realistic fantasy	No
style	hassaku		anime .base models hentai	Hassaku (hentai model) v1.2 by Ikena Hassaku, formerly known as Grapefruit, is an essential anime/hentai model with crisp, clean looks. It aims to be a model with a bright, clear anime style focused on nsfw images, but also with a high emphasis for good looking sfw images. This update improves eyes, hands, cleans up images, and reduces sepia tones. Check out our full walkthrough with prompt examples and tips to use Hassaku like a pro: https://www.animemaker.ai/hassaku-guide/ This update removed the 'black box' issue some users were experiencing.
inpainting	hassaku-inpainting	No		No
style	hassan14		realistic .base models semi-real porn	No
style	hassan1512		porn realistic .base models	The classic NSFW base model Hassan' Blend 1.5.1.2 is now retro by realistic porn standards, but by one community request we brought it back. It is primarily used for portraits, but it can also do basic poses like blowjobs and some BDSM. Created by Hassan, he notes "The NSFW content trained into it includes both male and female, various poses, scenes, types of shots and anatomy both up close and in various positions." Add /clipskip:1 to your prompt for best results
anime	hasumi		characters anime female	No
clothing	haute-couture		clothes	Haute Couture by EDG is a LoRA focused on high-class, luxurious dresses, a little on the skimpy side. Suitable for any kind of art base, from anime to realistic. A weight of 0.85 or less is recommended by the creator
lora	hd-porn-style	No		No
lora	head-pat		poses memes	No
effect	hearts-pov		anime hentai effects	No
inversion	heathered		characters female realistic white	This textual inversion creates the likeness of ASMR streamer, social media influencer Heatheredeffect Recommended settings for getting good renders: # Guidance at approx 7.0+ # Best weight around 1.0 # Works well non-realistic and realistic models # works well with basic and elaborate prompts Sidenotes: # Works well with other loras # Render at low sizes and use /more /size /strength to upscale Example Prompt: /render 1girl, solo, heathered, <heathered> [blurry, lowres, worst quality]<photon>
sci-fi	helghast		gaming sci-fi	This lora was designed and created by community member Eggs Benadryl. This lora is designed for creating realistic helghast soldiers from the Killzone video game series. It can produce high-quality images of the masked, orange-eyed enemies with minimal input. It works well with realism models. Works well with default weight.
style	henmix		realistic .base models semi-real	No
lora	henry-zebrowski		characters white male realistic	This lora, created by a community member, was designed to put podcast and sketch comedian Henry Zebrowski in your renders. It was found that higher weights help render quality as well as lower guidance around 7. Weights as high as 3 were acceptable.
style	hentai		anime .base models hentai	No
lora	hi3rd-elysia	No		No
clothing	high-fashion		clothes	No
lora	hila-klein		characters realistic female	A realistic likeness of model Hila Klein, by model creator Nomono. Works well with a base model like <believe> or similar
effect	hipoly		effects detailers	Add it with high strength to boost the 3D quality of your image, such as skin, hair, and clothing materials
clothing	hl2soldier		gaming characters clothes armor	No
style	hobbes		comics cartoons art style	No
sci-fi	holo-mech		anime mecha sci-fi armor comics	Hologram Mecha by Cinsdia Very cool Lora with hologram mechas! Recommended for Mecha: anylora model, steps: 20, CFG scale: 7, clip skip: 2 No triggers needed! This model makes some cool looking Mechs, but don't pigeonhole it by thinking that is all it can do, It also makes some REALLY nice fantasy armor too.
style	holokuki		.base models anime illustrations	Holokuki v2 by Aotsuyu Holokuki is a base model anime checkpoint merge, Project AI spinoff that aims to strike a balance "between detailed backgrounds and high-quality waifus", with special focus on V-Tubers. Judging from the comments on Civitai, it does a solid job on nsfw content
style	homoerotic		realistic .base models realistic male penis porn lgbtq	Homoerotic v2 by H4RV3YD3NT / gaydiffusion For the boys - Homoerotic V2, an 18+ gay men only concept. Here are some quick notes from the creator on how to use it, visit their Civitai page for more info and diagrams: "It' responsive to booru tags, types of shots ('medium shot', 'closeup', 'long shot', etc) as well as sentences/captions. Describing more aspects of the image you want will give you better results. Recommended quality prompt (at the least): masterpiece, best quality, highres, realistic Recommended negative prompt: black and white, monochrome, comic, bad anatomy, bad hands, text, error, cropped, letterbox, jpeg artifacts, signature, watermark, username, artist name, censored, 1girl, 3d rendering Certain body types may require some prompt alchemy. Try placing muscular tags in your negative prompt for less fantastic physiques. 'beefy' has always had a good effect as well as 'twink', both are self-explanatory and for some reason recognized by CLIP/BLIP. A sample prompt: masterpiece, best quality, highres, realistic, greyscale, monochrome, hat, male focus, photo background, navel, pants, topless male, nipples, street, outdoors, short hair, city, road, looking at viewer, police, jeans, pectorals, denim, twink [[black and white, monochrome, comic, bad anatomy, bad hands, text, error, cropped, letterbox, jpeg artifacts, signature, watermark, username, artist name, censored, 1girl, 3d rendering]]
anime	honey-bee		black characters anime female retro	Honey Bee aka Black Dynamite, an anime Lora that works with <av45> or similar. A default weight or less than 1 is recommended. use the whole set of triggers for the signature look, or leave them out for a general ebony afro female character
anime	honey-lemon		anime characters female	Honey lemon from Big Hero 6 character Lora Works with semi-real to anime pretty well. A weight somewhere around 0.7 works well with dreamshaper. Example prompt: /render /images:3 /size:1080x768 (Honey Lemon from big hero 6,), <honey-lemon:0.7> <dreamshaper>
pose	hoodie-underboob		female clothing poses	This lora creates images of women wearing short cropped hoodies Recommended settings for getting good renders: # Guidance at approx 7.0+ # Best weight around 0.4 - 0.5 # Works well non-realistic and realistic models # works well with basic and elaborate prompts Sidenotes: # Seems to interact well with other loras # Render at low sizes and use /more /size /strength to upscale. Example Prompt: /render 1girl, solo, oversized cropped hoodie underboob <hoodie-underboob:0.45> [worst quality, lowres, blurry] <photon>
clothing	hooters		effects clothes memes environments	No
pose	hotsquat		poses porn	No
style	howls-style		art style anime environments	No
anime	hsr-herta		characters female anime	No
anime	hsr-march7th		female characters anime	No
pose	hug-and-kiss		poses	No
pose	hug-and-suck		porn poses oral hentai	Hug and Suck concept by Emberbyte is "where the performer hugs the legs while they do their work". Bonus: community members soon noticed that if you prompt 2girls at low weights its becomes a passionate lesbian kiss lora when used with some anime models
lora	huge-assets		butts boobs anatomy	No
anime	hyper-bulge		hentai memes porn penis anime	Hyper Penis Bulge shows the outline of a giant dick inside of a stomach, as seen in hentai tentacle porn and wild movies. Needs low weight to work and not every base model is compatible. To get you started, try a prompt like this: /render a woman <hyper-bulge:0.1><aohard>
effect	hypermuscle		anatomy effects	HYpermuscle v1.0 by NOUsagi Muscles! Bigger and smaller. If you want them, use this lora. Add extra prompts like 'fit', 'well-build' etc to make the effect stronger. IMPORTANT! Works only with very low weights, from about -0.05 to 0.05
style	icomix		anime comics .base models cartoons illustrations art style	A higher resolution improvement blend of the original Comics Diffusion, however tends to be more biased towards women and anime. While this version has more of a 2D look, the newer blend of it <comicbabes> sports similar visuals with a softer digital painted comic look. Both are terrific and easy for beginners
style	icomix5		anime comics .base models cartoons illustrations art style	iCoMix v5 by lostdog Easy to use comics model. Version 5 is less 'cutesy' than previous ones and better fulfils its role as a comics model. The colours are also better and less grainy. Negative prompts that work well with it: (worst quality:1.4), (low quality:1.4), malformed hands, bad hands, mutated fingers, extra limbs, blurry, bad anatomy, bad proportions, gross, disfigured, jpeg artifacts, text, child, depth of field Minimal version: (worst quality:1.4), (low quality:1.4)
style	icons		illustrations art style graphic design	Colored Icons / Logo Generator creates crisp vector style images when its <icons> trigger is used with <dreamshaper> at a weight between 0.8 - 1, and the size of the image is smaller than 600x600, then use /facelift /anime to boost the quality. Example: /render an icon of a pirate <icons><dreamshaper>
style	identity		anime creatures female .base models monster girls furry	IdentityCrisis v.EroDemonMix0.3ws by xerminator13 A really nice and easy to use model that' good for anime girls, monster girls, and furry girls. It' the best model for creating monster girls we've tried. Include some of the trigger words into your prompts for better results.
lora	idris		characters male black	No
style	illuminati		realistic .base models semi-real fantasy	Illuminati is a dark semi-realistic model intended for moody portraits and fantasy use. Surprisingly, this model was not trained on nudes and handles nsfw very poorly, so it' strictly grim reapers and superheroes here. Try it with a very low guidance of 2.5 for best results. The model has suffered bad reviews due to the politics surrounded its licensing and spat with another model creator, and it' over-use of three textual inversions to get anywhere near the example images. It' a mixed bag, but it' cool... when it works
pose	imminent-penetration-pov		poses anime porn penis	No
lora	in-the-shower	No		No
lora	incredibles-mom		female characters cartoons	No
inversion	indian		indian anatomy realistic	No
lora	indiandoll		realistic realistic indian anatomy	This Lora was made with the goal of creating beautiful, realistic looking Inidan women. Author recommends a weight: 0.2 & 0.3 for photos, 0.6 for paintings. Keep 'brown eyes' in the prompt to make eyes look better. Add 'bindi' to negative prompt to remove the spot on the forehead. Important to keep guidance low (6ish) to avoid colored blotches on face. Example base prompt: /render /guidance:6 a pretty indian model with brown eyes <indiandoll:0.25> <edge> Example negative prompt: [[photozoov15, cartoon, illustration, painting]]
style	ink-punk		anime .base models art style environments art style interiors architecture	No
lora	insertions-big		anatomy penis pussy	Generates big penis, and stretched vaginas. Simple enough to get working. Somewhere around 0.6 weight. Works with all styles pretty well.
lora	inside-creature-vore	No		No
effect	instajizz		effects porn hentai	No
lora	internet-explorer-chan	No		No
style	invincible		anime art style	This Lora is used to create reproduce the style of the Invincible animated show. Author recommends a strength between 0.7 and 1. Cannot do masked characters well and may require /facelift /anime to fix faces.
lora	irene		characters realistic female asian	No
lora	ithlinni		asian characters female realistic	No
lora	iu		characters realistic female asian	No
lora	ivanka		characters female white	No
object	ivorygold		effects environments objects architecture armor	Ivory Gold is an effects lora that turns objects, buildings, clothing, armor, and just about any structure into gold and white, giving it a rare and luxurious look. It' easy to use and works well at default strength for objects, but set it lower when making people. You'll know when the weight is too high when people' facial expressions get weird. Use it with concepts like <deliberate2> and <revani>
lora	jackal-nikke	No		No
lora	jacknicholson		characters realistic male white	No
pose	jackopose		poses hentai butts	No
pose	jackopose2		poses hentai	No
anime	jacksparrow		characters realistic anime male white	No
lora	jadey			This lora specializes in creating images of twitch streamer and influencer jadeyanh. It is intended to produce realistic and diverse portraits of her. The lora focuses on her visual appearance
lora	jane-fonda		memes characters female white	No
lora	japandoll		characters realistic asian anatomy	No
lora	japandoll15		characters realistic asian anatomy	No
clothing	japanese-school-uniform		japanese asian clothes	No
style	jar-girls		effects anime semi-real art style creatures objects fantasy female	No
lora	jasonbateman		characters realistic male white	No
lora	jav-ai		jav porn asian characters female japanese	No
lora	jav-aimi		jav porn asian characters female japanese	No
lora	jav-aki		jav porn asian characters female japanese	No
lora	jav-anri		jav porn asian characters female japanese	No
lora	jav-eimi		jav porn asian characters female japanese	No
lora	jav-hitomi		jav porn asian characters female japanese	No
lora	jav-julia		jav porn asian characters female japanese	No
lora	jav-kana		jav porn asian characters female japanese	No
lora	jav-koharu		jav porn asian characters female japanese	No
lora	jav-koizumi2		japanese asian characters female jav porn	No
lora	jav-lemon		jav porn asian characters female japanese	No
lora	jav-lori		jav porn asian characters female japanese	No
lora	jav-meguri		jav porn asian characters female japanese	No
lora	jav-mio		jav porn asian characters female japanese	No
lora	jav-moe		jav porn asian characters female japanese	No
lora	jav-momo		jav porn asian characters female japanese	No
lora	jav-nene		jav porn asian characters female japanese	No
lora	jav-rei		jav porn asian characters female japanese	No
lora	jav-remu		jav porn asian characters female japanese	No
lora	jav-rion		jav porn asian characters female japanese	No
lora	jav-riri		jav porn asian characters female japanese	No
lora	jav-saeko		jav porn asian characters female japanese	No
lora	jav-saika		characters female realistic jav porn asian japanese	No
lora	jav-saika2		jav porn asian characters female japanese	No
lora	jav-sakurai		jav porn asian characters female japanese	No
lora	jav-shoko		jav porn asian characters female japanese	No
lora	jav-sonata		jav porn asian characters female japanese	No
lora	jav-toko		jav porn asian characters female japanese	No
lora	jav-tsumugi		jav porn asian characters female japanese	No
lora	jav-yua		jav porn asian characters female japanese	No
lora	jav-yui		jav porn asian characters female japanese	This Japanese Adult Video star can best be rendered with a low weight (0.05-0.3) and a complimentary base and style. For example, a good prompt could be: /render a pretty woman <bra4><jav-yui:0.15> #fuji
celebrity	jconnelly		characters female white	A LoRA in the likeness of actress Jennifer Connelly, by model creator AiNow. Tested with <realvis13> at a weight of 0.8
effect	jellyfish-forest		effects anime illustrations paintings	A fantasy LoRA effect by model creator L_A_X, Jellyfish Forest creates a background of glowing jellyfish in a forest setting, creating a magical and beautiful scene. The model is based on the base of Niji, so similar anime base models like <aom3> also work well. It' also possible to make the jellyfish appear in urban areas and floating over water
celebrity	jennette-mccurdy		female characters realistic	a Lora that produces photos inspired by the actress/author Recommended settings for getting good renders: # Guidance at approx 5.0 - 7.0 # Weight is best between 0.6 and 1 # Works great with realistic models # Works great with basic and elaborate prompts Sidenotes: # Seems to interact well with other loras # A good resolution to render at is /size:640x640 & /size:640x832 Basic Prompt: /render <jennette-mccurdy:0.6> portrait photo of "jennette mccurdy" [blurry, lowres, worst quality] <level4>
lora	jennifer-connelly		characters female white	A LoRA in the likeness of actress Jennifer Connelly, by model creator AiNow. Tested with <realvis13> at a weight of 0.8
inversion	jennifer-connelly-80s			No
celebrity	jenniferloveh		characters realistic female white	No
celebrity	jenny-ortega		characters realistic female latin	No
celebrity	jessica-alba		characters female realistic	No
celebrity	jessica-henwick		asian characters female	Jessica Yu-Li Henwick is an English actress, writer and director. She began her career in the BBC series Spirit Warriors and Silk. She then played Nymeria Sand in the HBO series Game of Thrones, Testor in the film Star Wars: The Force Awakens, and Colleen Wing in the Marvel series Iron Fist
anime	jibril		anime characters female	No
effect	jimlee		art style effects comics cartoons marvel	This lora gives your renders the art style of Jim Lee. While it can make characters like Superman, Batman, etc, remember that it' not focused on those. The author suggests a weight between 0.65-1. Use the prompt 'jim lee' for better effect.
lora	jingtian		characters female asian chinese realistic	Sally Jing Tian (Chinese: ??, born 21 July 1988) is a Chinese actress. She is known for her roles The Warring States, Police Story 2013, and Pacific Rim Uprising. This very flattering likeness of her is perhaps more beautiful than the actual person... but who' complaining?
lora	jmilf		characters female japanese asian anatomy	No
lora	johncena		characters realistic male memes white	No
lora	jordynjones		characters realistic female	No
effect	jp-idol		asian effects japanese characters female	A general beauty/effects LoRA for creating portraits of Japanese Idols, meaning young popular models. This is version 10, designed for <chilloutmix> base, works with a weight of 1 or less. Add the extra trigger to simulate a magazine cover, though note that the text will be random gibberish
lora	judy-hopps	No		No
style	juggernaut		.base models realistic	No
lora	julie-bowen		characters female	Julie Bowen Luetkemeyer is an American actress. She is best known for starring as Claire Dunphy in the ABC sitcom Modern Family, for which she received critical acclaim and six nominations for the Primetime Emmy Award for Outstanding Supporting Actress in a Comedy Series, winning in 2011 and 2012
anime	juniper		characters female anime hentai monster girls	No
lora	k-berry		female characters korean	a lora that creates the likeness of a famous Korean Twitch streamer/Maxim model Recommended settings for getting good renders: # Guidance at approx 7.0+ # Best weight right around 0.65-1.0 # Works best with realistic asian models but can have interesting results with non-realistic (recommend:<majic6>, <chilloutmix>) # Works great with basic and elaborate prompt Sidenotes: # Seems to interact well with other loras # Render at low sizes and use /more /size /strength to upscale # She currently has blonde hair. If you want to reflect that use a heavy weight in your prompt (blondehair:1.9) Example Prompt: /render /sampler:dpm2m /guidance:5 /size:512x768 1girl, solo, a beautiful Korean model, <k-berry:0.8> [childlike, exaggerated features, blurry. lowres, worst quality> <majic6>
lora	k-yeonwoo		korean asian characters realistic female	A lora inspired by the Korean Twitch streamer/adult model. Can be used at full weight. Seems to work best with the model <majic>. Example Prompt: /render /sampler:dpm2m a beautiful Korean woman on the beach, tiny red micro bikini, [blurry, lowres, worst quality]<k-yeonwoo><majic>
lora	kanna-hashimoto		japanese asian characters female	No
lora	kate-middleton		characters female white	Catherine, Princess of Wales GCVO is a member of the British royal family. She is married to William, Prince of Wales, the heir apparent to the British throne
lora	kathwin		characters realistic female white	No
inversion	katy-perry		characters female realistic white	No
anime	kemono		anime .base models furry	By request, Cross Kemono - a high quality anime model focused on furry characters in a paint and watercolor style that seems to excel in sports situations, such as high school gyms and asian town settings
lora	kevin-hart		characters realistic male black	No
sci-fi	khalai-mecha		anime mecha sci-fi fantasy armor comics	Khalai Mecha by Cinsdia Recommended for Mecha: anylora model, steps: 20, CFG scale: 7, clip skip: 2 No triggers needed! This model makes some cool looking Mechs, but don't pigeonhole it by thinking that is all it can do, It also makes some REALLY nice fantasy armor too (even better than Mecha). Check out this fantasy Mageward prompt from a lora user and prepare to be blown away... /render /size:512x768 /steps:30 /sampler:dpm2m /guidance:7 1 girl as a Mageward, best quality, masterpiece, ultra high res, (photo realistic:1.4), <khalai-mecha:0.9> [[easynegative, (nipple:1.4), logo, watermark, text, (worst quality:1.5), (low quality:1.5), (normal quality:1.5), low res, bad anatomy, bad hands, normal quality, black and white photos,(monochrome), (grayscale), cartoon,anime,drawing,painting,manga]] <revani>
lora	kirara-genshin	No		No
lora	kirby		characters gaming retro	This lora will let you play around with beloved video game character kirby. Default weight and anime models appear to work the best.
lora	kisumi-amau		characters female realistic	No
style	kiwi-style		art style anime	No
anime	klara		characters female anime gaming boobs	No
lora	knolling	No		No
lora	kobayashi-san-style	No		No
style	koenokatachi		anime art style characters	Lora based on the popular anime Silent Voice by Kyoto Animation. Great effect when used alone, but it can be easily mixed with other loras to obtain this specific art style. Works well with any weight.
style	koji		anime .base models illustrations	Koji v2.1 by Ikena <koji> is a more SFW version of hassaku. A great choice to use when you like <hassaku>' art style but would like to render something other than boobs. It still can generate NSFW if prompted. It was trained on Danbooru tags and reacts well to them. The author recommends using it with: "masterpiece" and "best quality" in positive "[(low quality, worst quality:1.4), monochrome, signature, text, logo]" in negative Use /parser:new /clipskip:2 and /sampler:dpm2m
inpainting	koji-inpainting	No		No
anime	koleda-belobog-zenless		characters female anime	No
anime	komi-shouko		characters anime female	Komi Shouko LoRA v 1.0 by Lykon # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low # Mangled limbs, general chaos ? Some LoRas can’t handle rendering with resolutions bigger than 512x512 // Also make sure your prompt isn’t too complicated for this LoRa # No visible effect ? Check if the LoRa has any triggers that need to be added in order to activate it # Good full models to try out LoRas with: anime ? <anylora> <counterfeit> <breakanime> comics ? <comics-vision> <icomix> <misstoon> semi-real ? <revani> <lyriel16> <dreamshaper6> realistic ? <majic> <bra5> <chilloutmix> <edge>
anime	komi-shuuko		characters female anime	Komi Shuuko v1.0 by mnemosynekotorin General lora guidelines: # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low # Mangled limbs, general chaos ? Some LoRas can’t handle rendering with resolutions bigger than 512x512 // Also make sure your prompt isn’t too complicated for this LoRa # No visible effect ? Check if the LoRa has any triggers that need to be added in order to activate it # Good full models to try out LoRas with: anime ? <anylora> <counterfeit> <breakanime> comics ? <comics-vision> <icomix> <misstoon> semi-real ? <revani> <lyriel16> <dreamshaper6> realistic ? <majic> <bra5> <chilloutmix> <edge>
lora	koreandoll		characters realistic asian anatomy korean	No
lora	koreandoll2		characters realistic asian anatomy korean	No
style	kotos		.base models semi-real anime fantasy	KotosMix V1, a semi-realistic model created by blending Protogen 2.2 (25%) and Abyss Orange Mix2 Hard (75%). The creator suggests short simple prompts for best results. This model became known for its must-be-seen results when you rpompt ((gigantic breasts)) if they even fit in your image all at once
lora	kpop-man		asian male korean characters	No
lora	kryten			Lora created by Community Member Eggs Benadryl Will generate the likeness of Kryten from the sci-fi comedy series Red Dwarf. Weights at or above one seem to work best. "Don't be a smeeeeeee"
style	kstyle		semi-real .base models porn asian korean	Korean 2.5D is a semi-realistic painted style focused on Korean female portraits, however it can also create males. It' an 18+ focused checkpoint model, try it as a base with asian character loras like <suzy>
style	kustom-kulture		vehicles art style	This lora is for creating kustom kulture artwork, a style of art that features exaggerated and cartoonish depictions of hotrods, motorcycles, and other vehicles. The lora can produce images that resemble the work of Ed "Big Daddy" Roth, the creator of Rat Fink, a popular character in kustom kulture. The lora can also generate original characters and designs based on the user' preferences and inputs. The lora is ideal for anyone who wants to create unique and eye-catching art inspired by the kustom kulture movement.
anime	lamia-lami		creatures anime semi-real female monster girls	Good LoRa for turning your favorite girl into a lamia. Easy to use and quite adaptable to different situations. It can be combined with the other lamia LoRa for a change in results. Weight: 0.4-1.2 Treat it as the legs factor. The lower the weight, the bigger the chance legs will appear. It worked well with any of the base anime and semi-real models it was tested with. The author recommends using it together with <aohard>. It works best when the size is increased to around 640x832 or higher. Using 512x512 resolution often causes distorted faces. This LoRa reacts well to such prompts as popular tail colors (blue, red, green, white, and purple) and hair colors. It was trained on many characters who had tops, dresses, or kimonos; those were also tagged as prompts.
anime	lamia-miia		characters creatures anime fantasy female	No
anime	lamia-suzu		creatures anime female semi-real monster girls	Good LoRa’s for turning your favorite girl into a lamia. Easy to use and quite adaptable to different situations. It can be combined with the other lamia LoRa for a change in results. Weight: 0.4-1.2 Treat it as the legs factor. The lower the weight, the bigger the chance legs will appear. It worked well with any of the base anime and semi-real models it was tested with. Try it out with <revani>, <counterfeit>, and <meinamix>. It works best when the size is increased to around 640x832 or higher. Using 512x512 resolution often causes distorted faces. Reacts well to prompts like popular tail colors (blue, red, green, white, and purple) and hair colors. It was trained on many characters who had tops, dresses, or kimonos, and those were also tagged as prompts. This LoRa may be used to create Miia and Jasmin-chan but isn’t too accurate with those.
lora	lanarhoades		characters realistic female white	No
lora	large-areola		boobs porn anatomy	Large Areolae (sic) by NoName_Sister can be controlled by setting the weight, where default looks like typical big boobs and 1.5 will more than raise your eyebrow. Tested with Dreamshaper7 but appears to be very compatible all around
lora	large-futanari		penis trans lgbtq porn hentai	No
inversion	latin	No		Latin/Latino Texual inversion aka Ethnic Helper by DevilKKW is a positive textual inversion. Use it like this: ((<latin:2>)) You can also use it wrapped in negative brackets to reduce Latin features, which will make your characters more pale with smaller eyes and straighter hair
lora	latinadoll		latin characters female	No
lora	laura-bailey		characters female white	Laura Bailey is an American voice actress and geek culture influencer, playing multiple anime characters for Funimation and dungeon master at Critical Role
lora	laura-matsuda-sf	No		No
lora	lauren-boebert		characters female realistic	No
lora	leg-panties	No		No
clothing	leggings		clothes	No
effect	lego-style		effects toys	No
style	lego-style2		toys art style	A LEGO toys LoRA, by konyconi. It was trained with <revani> as the base art style, though other models also work. The creator recommends /clipskip:2 and a guidance scale of 7 at /steps:60
lora	legs-together			No
pose	legs-up		poses porn hentai	No
pose	legs-up2		porn poses butts	No
lora	leilei-darkstalkers	No		No
inversion	less-asian	No		Style Less Asian by Zovya reduces the anime and asian properties of an image. The creator says "it doesn't race-swap, it just removes general asian influence so your prompts can be more effective with whatever race or culture you're trying to portray". This is a negative, so use it like the example below. It doesn't fix anatomy so pair it with something like very bad negative: [[<less-asian:-2><verybad-negative:1>]]
base	level4		realistic .base models semi-real fantasy paintings	HOT ONE! Great for beginners. Real / Semi-realistic general model for portraits, cars, landscapes. The creator demos it as the go-to model for creating female athletes. Try the #cinematic recipe with it
style	leyendecker		illustrations art styles	Joseph Christian Leyendecker was one of the most prominent and financially successful freelance commercial artists in the U.S. He was active between 1895 and 1951 producing drawings and paintings for hundreds of posters, books, advertisements, and magazine covers and stories. This lora is intended to mimic his artistic style.
style	liberty		realistic .base models semi-real fantasy porn	A mega-merge of 23 models. Liberty looks like photos taken from a cell phone, flawed, natural looking pictures of real people. Is NSFW and very short prompts can do wonders, don't overcomplicate it. Use trigger words in column H for best results (not at once)
lora	lick-gun	No		No
lora	licking-oral		oral porn	No
effect	light-and-silhouette		effects lighting art style environments	No
effect	light-shadow		effects lighting art style environments	Introduces light and shadow effects to your image. Recommend to start with strength of 0.4 - 0.8 and change accordingly. Does not need trigger words, but suggest using combination of the following prompt words like starry, starry light, night, colorful, cloud, star sky
inversion	lightning	No		No
lora	lilith-darkstalkers	No		No
celebrity	lily-collins		characters female	This lora is designed to create artistic images of Lily Collins, a famous actress and model. It can capture her distinctive features, such as her eyebrows, eyes, and smile, in various styles and settings. Whether you want a realistic portrait or a cartoon-like illustration, this lora can produce high-quality results that match your preferences
inversion	lindsey-pelas		characters female realistic white	Lindsey Pelas by JernauGurgeh is a positive textual inversion, a likeness of an American model. Add <lindsey-pelas> to the positive prompt to use it with your favorite base model
effect	liquid		effects	No
lora	liuyifei		characters female asian chinese	Crystal Liu, better known by her stage name Liu Yifei, is a Chinese-American actress, singer, and model. She has appeared multiple times on Forbes' China Celebrity 100 list and was named one of the New Four Dan actresses of China in 2009. She is widely known as "Fairy Sister" in China, and was the lead role in Disney' live action version of Mulan
inversion	livvy-dunne		characters realistic female semi-real	Olivia Livvy Dunne artistic likeness by JernauGurgeh, intended for use with <reliberate> For a more realistic Olivia, check <olivia-dunne> LoRA in our system, or combine the two with low weights for an effect somewhere inbetween
celebrity	Lluna-C		characters female realistic	a likeness of Spanish streamer Lluna C, created by Madridpool Joe
lora	loba-apex		characters semi-real female gaming	No
style	lofi		realistic .base models realistic	Lofi V2 is a unique model focused on color accuracy and different DSLR cameras, like a high-res digital camera version of <analog>. Works great with outdoor photos. Note: Many users have noted that it skews towards asian women without strong prompts for men and European faces. Can also do paint/semi-real and celebrities. Try these exact palettes: yellow, red, green blue, cyberpunk. For different bokeh styles try these four camera types (but not together) and spell Canon with this typo: Canan EOS R6 Mark II, Canan EOS M6 Mark II, SONY FDR-AX60, Panasonic WXF1
effect	logh-fpa-fleet		effects sci-fi environments objects spaceships	This LoRa art style resembles that of an old school japan sci-fi anime 1988 - Legend of the Galactic Heroes. It’s focused on the starships appearing in this series. Using the optional prompt “ginga eiyuu densetsu” may result with better results. The recommended weight is 0.3-0.7, and guidance 6-8.
style	logo-modern		art style graphic design	No
style	loish		anime paintings female art style	Loish is a paint style loosely modeled after the blushing women painted by Lois van Baarle. The style is something like a cherub-faced Sam Does Arts, with curvy short anime women. It will frame images in a medium cowboy shot by default. Tested with <aohard> and <perfectworld> at guidance 7 and a weight of 1, it takes some wiggle to get a sharp image. Not recommended for beginners. Try it with /facelift /anime for sharper results
anime	lol-gwen		characters female gaming anime	Gwen League of Legends \| 2 Outfits \| Character Lora 870 v1 by Numeratic Lora trained on 870 images of Gwen from league of legends. It does not like her scissors. Use those prompts if needed: "holding weapon, holding, scissors, oversized object, holding scissors" Recommended weight: ~0.25-0.4 Use one of the triggers for different clothing options.
lora	lol-jinx		characters cartoons gaming	a lora that creates renders of the character Jinx from LOL Recommended settings for getting good renders: # Guidance at approx 7.0+ # Best weight right around 1 # Works great with realistic models and non-realistic models # Works great with basic and elaborate prompts # This is an excellent lora that performs great right out of the box, VERY robust Sidenotes: # Seems to interact well with other loras (for fun try it with the <gun-point> lora) # Render at low sizes and use /more /size /strength to upscale # Some great renders can be created by starting with a realistic model like <level4> or <believe> and then remixing with your fave anime/comic model like <toonyou>. Can produce an almost cel-shaded effect. Experiment!! Example Prompt: /render 1girl, Jinxlol, <jinx-lol> [blurry, lowres, worst quality] <revani>
lora	lorengray		characters realistic female white	No
lora	lost-depth	No		No
effect	lowra2		lighting effects	LowRA v2 by XpucT is a Low-light LoRA, hence the name. The recommended weight range is 0.6 to 0.8, so it is set to 0.7 by default. For additional darkness, add the words "dark theme" to your prompt (without the quotes). It was designed for the base model <deliberate2> but works with other models, too
style	lucidmix		semi-real .base models fantasy environments	No
lora	lucoa-dragon-maid	No		No
celebrity	lucyliu		characters realistic female asian	No
style	luis-royo		art style fantasy	No
celebrity	lyasyaa		female characters realistic white	A realistic LoRA of the model resembling streamer on twitch, Lyasyaa, created by MarkBW. Works well with <galaxytime>, <deliberate2> and <realvis20>. Author' comment: "Trained on v1-5-pruned with 30 pictures. Best result with weight between : 0.3-0.4. Other prompts : long hair, brown hair, blonde hair, brown eyes, lipstick, makeup, realistic, red lips. Negative prompt : green eyes."
style	lyriel		semi-real .base models anime environments architecture interiors	A moody semi-realistic fantasy BASE model, generally designed for portraits and full-length anime style photos. Fantastic landscapes and food photos, doesn't require long prompts. Try your favorite prompts from Deliberate, especially food! Try it with #cinematic or #easynegative
style	lyriel16		.base models semi-real fantasy sci-fi	Lyriel version 1.6 (updated May 26th) is an iterative update to the already excellent Lyriel 1.5, a digital painting base model with a distinct artistic style. Its a general model but is best known for dramatic portraits and strong results with just short prompts. The creator recommends using /sampler:dpm2m WHAT'S NEW: -More realistic characters and environments, Shadows and lighting, quality of food, and detailing has been improved
effect	m-pixel		anime effects cartoons characters	LoRA to make pixel characters. Recommended weight to use is from 0.1 - 0.3, increasing the weights makes the pixel character/background smaller. Use anime or semi-real bases. Token words: pixel Trigger words: pixel, pixel art For drawing characters, recommend to use size: 512x768, weight = 0.2. For drawing buildings, recommend to use size 768x768, weight = 0.2
inversion	macro-giant		effects	Macro by DevilKKW adds a giant subject to the image. Use it in your positive prompt
celebrity	madonna		characters realistic female retro white	No
clothing	maid-bikini		clothes	No
clothing	maid-dress		clothes anime	An anime-trained model for a maid cafe costume or maid' sexy dress for role play. Use an anime base model with a weight between 0.6 and 0.8 for best results. The model was trained with NAI by NicerPanda
inversion	maitland-ward			This textual inversion creates the likeness of former tv star, adult film star Maitland W Recommended settings for getting good renders: # Guidance at approx 7.0+ # Best weight around 1.0 # Works well non-realistic and realistic models # works well with basic and elaborate prompts Sidenotes: # Works well with other loras # Render at low sizes and use /more /size /strength to upscale Example Prompt: /render 1girl, solo, maitward, <maitland-ward> [blurry, lowres, worst quality]<photon>
style	majic-horror		chinese .base models asian fantasy horror	majicMIX horror v.1 by Merjic <majic-horror> is an unexpected offshoot of the majic series that produces fascinating horror-like renders. While they're not as creepy as the good old <creepy>, the quality of the renders is much better. It can add a dark spin to every pretty woman generated by the standard majic model. Tips: - To obtain more varied and less human-like results, use /parser:new /clipskip:2 - It works great with /sampler:k_euler_a - Reducing the number of steps to 20 or 30 can bring out wilder results, since they won't have as much time to get 'prettied up' - Recommended prompts: horror art, dread, creepy, bloody, gore ... everything about horror can be explored. - Try it together with <body-horror>, <cyberdead>, <nightmarish>, <giger> or other loras from the horror category
style	majic6		chinese .base models asian realistic	majicMIX realistic v6 by Merjic Majic Realistic Version 6 is a remake of version 5 since the author decided that 5 isn't realistic enough. It' great at creating beautiful Asian portraits. Similar to Chilloutmix, it can be used as a base to interact with LoRas. A few tips: # It doesn't need complicated prompts, the most natural look is reachable with simple prompts. # Recommended settings by the author: /sampler:k_euler (remove '\') /Steps: 20~40 /guidance:6-8 /parser:new /clipskip:2 # Recommeneded prompts: - positive prompts: Best quality, masterpiece, ultra high res, (photorealistic:1.4), 1girl - negative prompts: [(worst quality:2), (low quality:2), (normal quality:2), lowres, watermark, monochrome] - IMPORTANT! If you use style or recipe remove the weight:2 or the quality may drop considerably. - f you want darker picture, add: in the dark, deep shadow, low key
celebrity	mana-ashida		characters female japanese asian	No
style	manga		art style anime illustrations	Nieta Art' Pen Sketch Style LoRA is intended to use with <av45> at a weight of 0.8 to create a monochrome manga-style look of pen sketch style japanese comic book illustrations
fantasy	manhua-comics		art style asian comics chinese manga illustrations manhua sci-fi fantasy	comic ???? v1.0 by sincity This Lora creates an Asian comic style called Manhua... Manga, manhwa, and manhua are often grouped together, but each of these comic forms has it' own unique features. While manga originated in Japan, manhwa originated in Korea, and manhua originated in China. These comics share many similarities but also have differences in art styles, storytelling, and readership. This Lora was trained on meinah3, and works well on majic5.
celebrity	marilyn-monroe		female characters	a lora that creates the likeness of Marilyn Recommended settings for getting good renders: # Guidance at approx 7.0+ # Best weights from 0.7 to 1 # Works great with realistic models and non-realistic models # Works great with basic and elaborate prompts Sidenotes: # Seems to interact well with other loras # Renders best at 640x640 or 640x832 Example Prompt: /render 1girl, solo, marilynmonroe, <marilyn-monroe:0.8>[blurry, lowres, worst quality] <edge>
style	marker-style		effects art style	This lora is intended to add a marker style. Made to give the impression that it was colored in with crayola style markers. This lora was made by the creator of Mistoon who did not provide many details about the proper use of this lora. Through testing it was found that it works best with anime models and lower complexity images. This model was found to require very low weights around 0.15-0.35. Works best with misstoon or alongside the <toon-style> lora.
lora	marrobbie		characters female realistic white	No
style	mars		realistic .base models semi-real realistic	A new photorealistic model "Mars Mix", version 0.2, is now available. It' a blend of Ares + Photoreal + Herb mix, with support for danbooru tags, so it can do wild anatomy while still maintaining a realistic look
lora	martymcfly		characters realistic male retro white	No
style	marvels		cartoons illustrations comics .base models art style	M4RV3LS & DUNGEONS is a full base model for standalone or LoRA use, featuring five different comic and cartoon illustration styles. Use the triggers in your prompt to modify the effect: comic style, illustration, line art, comic poster, and art styles like Chiaroscuro can also be good modifiers. You can add "realistic" as a tag but the photo won't turn real, it will just look more shaded, more like semi-real
effect	masterpiece		anime detailers effects	Masterpiece is a detail-boosting LoRA by Eternal2kPP that boosts the overall aesthetic quality of an image by preferring design elements commonly found in concept art and promotional materials, where the character and settings have ideal framing and lighting. It isn't necessarily an anatomy booster, but it can do that as well. It may change the look of the character and the pose, so you'll want to adjust the weights carefully. Try it with a weight of 0.7 and your favorite anime or hentai base.
pose	mating-press		poses hentai porn	This is a LoRA to generate Mating Press, which is the male dominant sexual position thrusting downwards while the camera focuses on the press-like action, usually with room to prompt for the facial expression of the person getting pressed. The creator suggests using this at 0.8-1 weight for best results but make sure to always play around with the weight. Recommended bases: <aohard><av3><aom3> It was trained on 512x512 and works best with this size. To make the results look better, use "/more /size:832x832 /strength:0.4 /guidance:16" or a variation of it. Fixing the aspect ratio to wide helps: If you make the picture wider - It will make the effect of deeper penetration. Yhuur!
anime	mecha		anime mecha armor comics oral. porn oral 3d	Mecha by Joe-Di Welcome to the Mecha model. It can produce great effects for both anime and 3D. For Anime, try breakdomain base model For 3D, use ReV Animated. Weight: 0.6 Sampler ?Euler a, and DPM++ 2M Karras. Each has slightly different effects, so you can test them out for yourself. Guidance? 7-9, Steps : 20-40.
clothing	mecha-karukan		mecha art style characters anime clothes armor	Introduces mecha-armor to your renders and karukan artist style depending on your purpose. Keywords: mecha to add mecha armor, karukan for the karukan artist style. Strength: 0 - 1
style	mecha-style		mecha armor art style cyberpunk	Mecha Style by ArienTOP Trigger Words : mecha This model is used to replace mecha skins for characters?Perfect for cyberpunk and other styles Recommends: Steps: 20, Sampler: DPM++ 2M Karras, Guidance: 7
sci-fi	mechamuse-a		objects mecha anime armor	No
sci-fi	mechamuse-c		objects mecha anime armor	No
lora	mechanical-horizon		sci-fi fantasy gaming art styles creatures vehicles effects	No
sci-fi	mechanical-parts		anatomy sci-fi armor	Real Mechanical Parts v1.2 by Reelai In addition to increasing the lora weight, you can also try adding prompt (mechanical parts:1.x) where x is equal to a number to increase number of mechanical parts on your model. I recommend it to set that number between 1.1 to 1.5. Add the trigger as well if you want even more mechanical parts.
style	meinah		anime .base models hentai	V 2.1 update - The hentai version of Meina Mix, nuff said. High quality super horny hentai model, but it' pretty colorful and crisp
style	meinah3		anime .base models hentai	V 3 update - The hentai version of Meina Mix, nuff said. High quality super horny hentai model. Compared to the previous version this one is darker and a bit less realistic. Generated images tend to be sharper as well.
style	meinamix		anime .base models illustrations	MeinaMix v10.0 by Meina One of the best-looking anime base models bar none, this is the "objective" Meina with a high-contrast vae produces vibrant, stunning anime images with very short prompts. A great one for beginners and pros will also find a lot to love here. Works well with #easynegative
style	meinap		anime .base models illustrations paintings	Meina Pastel is a full model, a great pairing for Paint style anime LoRAs. It is a variant of the more traditional anime base model, MeinaMix. This version features looser, chunky edges and produces a wider range of colors in random and unexpected patterns. Check our full guide and prompt examples to learn to use it like a pro: https://www.animemaker.ai/guide-1-meina-pastel/
style	meinart		anime .base models illustrations paintings	Meina Pastel v5 by Meina <meinart> is a full model, a great pairing for paint-style anime LoRAs. It is a variant of the more traditional anime base model, MeinaMix. This version is an artistic variation that doesn't delve so much into the painting territory as <meinap>. Nevertheless, it excelles at SFW original renders that can be compared to detailed anime illustrations.
style	memphis-style		retro art style effects	This Lora, designed by a community member is intended to emulate the style of the famous 1980s Memphis Group. Weights about 0.3 are recommended as memory errors are likely any higher.
anime	merchant-dragon-quest-3		characters female boobs anime gaming	A nsfw interpretation of the merchant from Dragon Quest 3 videogame, it was trained using the Kotos base model so it is overflowing with boobness. Naturally, combine with <kotos> for the intended result
anime	metalslug-fio		gaming characters female asian anime	The female nerdy character from the videogame Metal Slug. Use an anime base like Anything V3 <av3> and a weight of 0.5 for best results. Various fun trigger options also available
sci-fi	metroid-samus		characters semi-real female gaming	Inspired by the heroine from Metroid, it' a likeness of Samus Aran without the helmet. Using a weight of 0.5 seems to be the sweet spot. Her mole is hit or miss but, if you bump the lora weight up to 0.6 or 0.7, it shows up more consistently; at the cost of slightly less defined lips and eyes. Semi-real base models like <perfectworld> work best, though you can use any kind. Notes from the creator, Itsame498: You can get some pretty realistic looking results, depending on your tags (obviously). Also, the Zero Suit seems to render accurately in most of my testing. Just use the (zero suit) tag. Her Paralyzer gun was in the training data. Using (gun) gets you in the ballpark, but I just can't seem to get weapons to render accurately. You may have better luck with some experimentation. Check the triggers for more options
anime	mgq-queenvamp		anime characters female monster girls	LoRA trained on Queen Vampire from Monster Girl Quest. Use trigger words mgq_queenvampire to create the character look, and use the following trigger words ringlets, black mantle, throne to create the environment. Recommended weight is from 0.1 - 0.4, where 0.1 - 0.2 creates weak effect of the LoRA. Stronger effect is at 0.3 - 0.4. Anything above 0.5 reduces quality
anime	mha-toga		characters anime female	Toga Himiko v1.0 by ThatstheguyIknow # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low # Mangled limbs, general chaos ? Some LoRas can’t handle rendering with resolutions bigger than 512x512 // Also make sure your prompt isn’t too complicated for this LoRa # No visible effect ? Check if the LoRa has any triggers that need to be added in order to activate it # Good full models to try out LoRas with: anime ? <anylora> <counterfeit> <breakanime> comics ? <comics-vision> <icomix> <misstoon> semi-real ? <revani> <lyriel16> <dreamshaper6> realistic ? <majic> <bra5> <chilloutmix> <edge>
anime	mha-yaoyorozu		anime female characters	No
anime	mhutiki		characters anime female	Mhu'tiki / Legend of Queen Opala / Cat girl / gyaru v.med by Tomas_Aguilar General lora guidelines: # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low # Mangled limbs, general chaos ? Some LoRas can’t handle rendering with resolutions bigger than 512x512 // Also make sure your prompt isn’t too complicated for this LoRa # No visible effect ? Check if the LoRa has any triggers that need to be added in order to activate it # Good full models to try out LoRas with: anime ? <anylora> <counterfeit> <breakanime> comics ? <comics-vision> <icomix> <misstoon> semi-real ? <revani> <lyriel16> <dreamshaper6> realistic ? <majic> <bra5> <chilloutmix> <edge>
inversion	mia-khalifa	No		No
clothing	michael-jackson		characters realistic male clothes black	No
lora	michaela-coel		black characters female	British Actress from Black Mirror, Michaela Coel
inversion	micro		effects	Micro by DevilKKW makes everything a miniature. Use it in your positive prompt
clothing	microshorts		clothes anime realistic semi-real	Clothing Lora which enhances/corrects the effect of microshorts. Use tokens microshorts. Recommended strength is from 0 - 1, your mileage may vary. Use denim shorts as either positive or negative prompt to add-on or remove with this lora. Works for anime, real and semi-real bases
lora	miko-lee		asian chinese porn characters female	No
lora	milana-vayntrub		characters white female	No
anime	milim-nava		creatures anime female characters	A female character likeness of a heroine in the TV show "That time I got reincarnated as a slime". Demon Lord Milim Nava, a dragonoid comes with two triggers for outfit changes, the default and random. It' a NSFW character lora, so use it with <aohard> or <aom3> at a weight of around 0.8
lora	milk-breast-pose	No		No
object	milking-machine		objects hentai porn	No
lora	milly-alcock		characters female	This lora renders the likeness of actress Milly Alcock (House of the Dragon) Recommended settings for getting good renders: # Guidance at approx 7.0+ # Best weight around 0.5 # Works well non-realistic and realistic models # works well with basic and elaborate prompts Sidenotes: # Seems to interact well with other loras # Render at low sizes and use /more /size /strength to upscale. Example Prompt: /render 1girl, solo, <milly-alcock:0.55> [worst quality, lowres, blurry] <photon>
lora	minvecmecha		armor comics mecha	Minimal Vector Mecha by Cinsdia Recommended for Mecha: anylora model, steps: 20, CFG scale: 7, clip skip: 2 No triggers needed! This model makes some cool looking Mechs, but don't pigeonhole it by thinking that is all it can do,This one also makes some old-style comic-book looks as well.
lora	mira004		characters female	This lora specializes in creating images of Twitch streamer Mira004. It is intended to produce realistic and diverse portraits of her. The lora focuses on her visual appearance. Use it at a weight of 0.4 for best results.
lora	miranda-nikke	No		No
lora	miss-martian	No		No
pose	missionary-anal-pov		poses porn	No
pose	missionary-pov		poses porn	No
style	misstoon		anime .base models cartoons	This model creates realistic and expressive characters with a "cartoony" twist. Unlike other anime models that tend to have muted or dark colors, <misstoon> uses bright and vibrant colors to make the characters stand out. It' ideal for anyone who loves anime and wants to create their own unique style or simply something different compared to other anime models.
style	mochizuki		anime illustrations art style	Lora based on the art of Mochiduki Ken. It has a strong effect and tends to produce results in either monochrome or sepia. It likes to add random pieces of text, but putting [text] in your prompts seems to cut down this lora’s training data significantly and lowers quality. Suitable for producing a base for remixing. Adding another character lora into the mix may positively affect the outcome, but <mochizuki> overwhelms them even when used at low weight.
style	modern-disney		semi-real .base models cartoons art style	No
lora	momofuki		anime hentai boobs butts anatomy illustrations art styles	No
style	monstergirls		anime creatures female .base models monster girls	All-in-One Monster Girls by Pjaama is base model for creating half-monster anime girls. It has special training for the words Lamia, Harpy, Centaur, Scylla, Mermaid, and Arachne, however any sort of monster should work. Also makes monster dudes. This is the full checkpoint train base model, so you can use it with LoRAs
pose	mooning		poses porn anime semi-real butts	No
lora	morrigan-darkstalkers	No		No
lora	motoko-kusanagi-sac	No		No
pose	motorbike-riding		poses objects vehicles	No
object	mouthgag		objects bondage hentai	No
style	moxin		chinese art style illustrations paintings asian	MoXin is a Lora trained from on Chinese painting Masters lived in Ming and Qing dynasties. It is recommended to use this with Chilloutmix or Guofeng as the base model, and note the five different triggers for variations in style. Model creator Simhuang recommends a LoRA weight of 0.7-1
lora	ms-fortune-skullgirls	No		No
anime	mt-ghislaine		characters female anime	Ghislaine (Mushoku Tensei) v1.1 by Monfor Salentaiel # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low # Mangled limbs, general chaos ? Some LoRas can’t handle rendering with resolutions bigger than 512x512 // Also make sure your prompt isn’t too complicated for this LoRa # No visible effect ? Check if the LoRa has any triggers that need to be added in order to activate it # Good full models to try out LoRas with: anime ? <anylora> <counterfeit> <breakanime> comics ? <comics-vision> <icomix> <misstoon> semi-real ? <revani> <lyriel16> <dreamshaper6> realistic ? <majic> <bra5> <chilloutmix> <edge>
effect	muertos		effects environments	With this lora I've found the ideal weight is from 0.2 to 0.7 with our system. Basically it applies Dia De Los Muertos style makeup (La Calavera Catrina style) of the Mexican tradition of the Day of the Dead.
pose	multi-pov		hentai anime effects oral poses	A well-made pov LoRa that can be used for both NSFW and SFW renders. After removing all sexually suggestive prompts, it will be a nice help for capturing one scene from many different angles simultaneously. As for its more natural use, this is how it works: It can generate three types of multi-pov sex scenes; each has a different trigger. Examples: # From behind: pov, multiple views, 1girl, sex from behind, standing, smile, black dress, completely nude, # Missionary: pov, multiple views, standing, smile, school uniform, head back, orgasm, # Fellatio: pov, multiple views, 1girl, fellatio, crouching, standing, smile, school uniform, head back, orgasm, Works best at either 512x768 or 768x768. Tall images give the overlay effect better, but square images give you better control over the different scenes. If you include completely nude and uniform on a square image prompt, you're a lot more likely to get that, with the left frame being completely without clothes and the right frame completely dressed. Works with most anime base models, but it isn't suited for more realistic ones. The author used <aom3> in his examples on civitai. Weight ~0.8-1 for normal use. On lower weights, it tends to generate less sexy stuff, which may be used in order to make dakimakura covers, etc. Examples: # NSFW # /render /seed:483279 /guidance:14 /sampler:dpm2m /size:768x768 pov, multiple views, 1girl, sex from behind, standing, smile, black dress, completely nude, large breasts, mature female, masterpiece, good quality, [lowres, blurry worst-quality, low quality, bar censor, censored, mosaic censoring, testicles] <multi-pov> <aom3> # SFW # /render /seed:191796 /size:768x768 /guidance:9 /sampler:k_euler_a pov, multiple views, a girl in a classroom, seated on a chair in a classroom, looking at the window, one view from the side and one view from the front, short hair, seems bothered, ((deep shadows, masterpiece, good quality, intricate details)), cute, detailed, ethereal, sharp lines, 8k, uhd, [[lowres, blurry, worst quality, low quality, NSFW, fused arms, bad hands]] <multi-pov> <counterfeit>
pose	multiview		gaming illustrations poses anatomy anime characters comics 3d	multiple views v1.0 by yinxian47 This Lora will take your described character and generate multiple views making it perfect for 3D modelers, Character Concepts, Turnaround Angles, D&D Character Sheets and more. This can help a lot in getting some character consistency, as you could Photoshop the image to take out the pose(s) you want and use them in remixes. It is recommended to use a simple white background and for better results it is best used in a wider image like 768x512, but it can still do a square like 512x512. Also works best with the trigger "Multiple Views" right at the beginning of your prompt.
effect	muscle-girls		effects anatomy female	No
lora	muscle-slider		anatomy effects	Muscle Slider by Ostis adds or reduces bulk depending on the weight. Negative weight removes muscle, and adding weight adds muscles! Works with realistic and illustrations too
lora	n64		gaming retro	No
style	nagi-style		anime paintings art style	No
clothing	naked-overalls		clothes porn	No
lora	nanami-madobe	No		No
sci-fi	nanosuit		comics gaming sci-fi armor	Nanosuit v1.0 by EDG Crisis was a great first-person shooter from a few years ago. The game had a Nanosuit you could wear, and this Lora captures the look of your main character perfectly. The higher the strength, the more it will look exactly like the suits and may not follow your other prompts. You can also use this to create different looks for both male and female character prompts. Try Lora strength around 0.35-0.55, these seemed to be the best overall setting.
lora	naruto-fuu		characters female anime naruto	No
anime	naruto-hinata		characters anime female naruto asian	No
lora	naruto-kayuga		naruto anime characters female monster girls	No
lora	naruto-konan		anime characters female naruto	No
anime	naruto-sakura		characters anime female naruto asian	No
lora	naruto-tenten		naruto anime characters female	No
anime	naruto-tsunade		characters female anime naruto asian	No
inversion	natalia-dyer		characters female realistic semi-real paintings	No
lora	natalie-portman		characters female realistic semi-real white	A realistic and semi-real Lora of actress Natalie Portman by EuroTaku. The creator recommends 360x500 then /highdef, and rendering at /guidance:6 with a base like <realvis20> at a weight of 0.5. Also works well with fantasy base models for superhero looks and costumes
style	ned122		.base models semi-real	NeverEnding Dream version 1.22 (installed May 26th) by Lykon is the model to use when DreamShaper is too fantasy, and AnyLora is too anime. NED fits snuggly between the two. This is a full base model designed for high compatibility with LoRAs. Here are some notes directly from the creator: NED is good at: Generating cosplay images Generating anime pictures Working accurately with character LoRAs Generating good looking people Generating realistic animals Generating images using booru-like tags
inversion	neg-feet		.negatives	Negative Feet V2 by Garland reduces the likelihood of extra legs and misshapen feet when added to your negative prompt. Usage: [[<neg-feet:-1.5>]] If you are still experiencing deformities, try adjusting the negative weight between -0 to -2. It can also be combined with other negative textual inversions use it with your favorite loras and base models
inversion	negative-hand	No		Negative Hand by Nerfgun3 is a hands anatomy fixer that unlike other negative prompts will not alter the overall characteristics of your character. Remember to wrap the trigger inside a negative prompt like this: [[<negative-hand:2>]]
inversion	negative-hands		.negatives	Negative Hand by Nerfgun3 is a hands anatomy fixer that unlike other negative prompts will not alter the overall characteristics of your character. Remember to wrap the trigger inside a negative prompt like this: [[<negative-hand:2>]]
clothing	negligee		clothes	No
anime	nekomiya-zenless		anime characters female	No
style	neoartcore		art style illustrations anime	No
sci-fi	neonpunk		art style lighting sci-fi	Neonpunk is exactly what it sounds like: neon futuristic lights and cyberpunk style aesthetics, by model maker Machine Delusions. Tips: -USE THE TRIGGER WORD. Helps push the style. -sometimes it can be biased towards certain things like women, characters, and center compositions. Use negative weights to get rid of what you don't want -there is an overarching "bloom" effect or a "haze". Simply adding "Bloom" to the negative prompt helps a ton!
pose	netorare		poses porn hentai	No
inversion	new-negative	No		No
sci-fi	nick-valentine		gaming characters sci-fi post-apocalyptic	This lora was designed and trained by community member Eggs-Benadryl. This lora is designed to generate Nick Valentine, a character from the Fallout series. Nick Valentine is a synthetic detective with a cybernetic face and a trench coat. He has a distinctive voice and personality. This lora can create realistic models of Nick Valentine with varying degrees of cybernetic detail. This lora requires a weight of around 1 to 1.5 to produce optimal results.
anime	nier-2b		anime gaming characters	MUST BE 0.0 WEIGHT Will not work at even 0.1! works beautifully at 0.0 though. Different types of 2B are possible with the triggers. MUST BE 0.0 WEIGHT
pose	nightclub-girls		female environments poses realistic	a Lora that produces photos of multiple females in a nightclub setting Recommended settings for getting good renders: # Guidance works # Weight is best between 0.8 and 1 # Was trained on <chilloutmix> but works with most realistic based models also works with un-realistic models # Works great with <edge> Sidenotes: # Use "multiple girls" in your prompt for multiple women # Seems to interact well with other loras including character loras (works great with <flashingtits> and <skirtlift> # A good resolution to render at is /size:832x640 # Try throwing in terms like "wild party", "dancing", "singing" etc Basic Prompt: /render <nightclub-girls>,nightclub (multiple girls:1.3),<edge>
lora	nightmare-barlowe		horror fantasy paintings art styles illustrations post-apocalyptic sci-fi	No
lora	nightmare-series	No		No
style	nightmarish		art style horror effects	No
sci-fi	nijiarmor		armor cyberpunk fantasy sci-fi mecha	NijiArmor LORA - suits / armors / mechas by troistr This LORA was trained with the capacity of creating various types of armor suits and mechas. It is heavily inspired by nijijourney, but can also do realistic images. Recommended LORA weight: 0.8-0.9. No prompt activator needed, but it was designed to react to diverse trigger words. Suit/robot type: android, cybersuit, exosuit, hardsuit, mechasuit, mecha Armor spec: shoulder armor, upper armor, full armor Accessory: helmet, headgear, headset, cape, cloak, visor, mechanical jaw Aditional triggers: dark skin, bulky, mechanical arm, mechanical arms, mechanical parts Important note: this lora was designed to create mechas without wings or any other kind of back volume - it will only happen if the checkpoint used overrides it.
anime	nikke-brid		characters female anime gaming	Brid (Nikke) v3.0 by UglyBoyPig # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low # Mangled limbs, general chaos ? Some LoRas can’t handle rendering with resolutions bigger than 512x512 // Also make sure your prompt isn’t too complicated for this LoRa # No visible effect ? Check if the LoRa has any triggers that need to be added in order to activate it # Good full models to try out LoRas with: anime ? <anylora> <counterfeit> <breakanime> comics ? <comics-vision> <icomix> <misstoon> semi-real ? <revani> <lyriel16> <dreamshaper6> realistic ? <majic> <bra5> <chilloutmix> <edge>
anime	nikke-dolla		characters female anime gaming	No
anime	nikke-fx-yuni		characters female anime boobs	FX Yuni from the Nikke game by UglyBoyPig. Use a hentai base anime model like Kotos for best results to capture her, well, specific anatomy but at a LoRA weight of 0.4 or thereabouts
anime	nikke-helm		characters female anime gaming	No
lora	nikke-jackal		characters female anime gaming	No
lora	nikke-miranda		characters female anime gaming	No
anime	nikke-pepper		characters female anime gaming	No
lora	nikke-poli	No		No
lora	nikke-quency	No		No
anime	nikke-rapi		characters female anime gaming	Rapi Nikke v1.0 by EugenR # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low # Mangled limbs, general chaos ? Some LoRas can’t handle rendering with resolutions bigger than 512x512 // Also make sure your prompt isn’t too complicated for this LoRa # No visible effect ? Check if the LoRa has any triggers that need to be added in order to activate it # Good full models to try out LoRas with: anime ? <anylora> <counterfeit> <breakanime> comics ? <comics-vision> <icomix> <misstoon> semi-real ? <revani> <lyriel16> <dreamshaper6> realistic ? <majic> <bra5> <chilloutmix> <edge> Helpful prompts: mature female, large breasts,yellow eyes, blonde hair,lips,black choker,chest strap ,red necktie, serious, black jacket,military uniform,black collared shirt ,beret,thighhighs, thighs,pleated skirt,red gloves,open jacket, from behind, ass, ass focus,
anime	nikke-soda		characters female anime boobs gaming	Soda \| Goddess Of Victory (NIKKE) v1.0 by Goofy AI # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low # Mangled limbs, general chaos ? Some LoRas can’t handle rendering with resolutions bigger than 512x512 // Also make sure your prompt isn’t too complicated for this LoRa # No visible effect ? Check if the LoRa has any triggers that need to be added in order to activate it # Good full models to try out LoRas with: anime ? <anylora> <counterfeit> <breakanime> comics ? <comics-vision> <icomix> <misstoon> semi-real ? <revani> <lyriel16> <dreamshaper6> realistic ? <majic> <bra5> <chilloutmix> <edge>
anime	nikke-volume		characters female anime gaming	Volume (Nikke) v3.0 by UglyBoyPlg # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low # Mangled limbs, general chaos ? Some LoRas can’t handle rendering with resolutions bigger than 512x512 // Also make sure your prompt isn’t too complicated for this LoRa # No visible effect ? Check if the LoRa has any triggers that need to be added in order to activate it # Good full models to try out LoRas with: anime ? <anylora> <counterfeit> <breakanime> comics ? <comics-vision> <icomix> <misstoon> semi-real ? <revani> <lyriel16> <dreamshaper6> realistic ? <majic> <bra5> <chilloutmix> <edge>
anime	nikke-yan		characters female anime boobs gaming	No
pose	nipple-tweak		boobs poses	a lora that creates nipple tweaking renders Recommended settings for getting good renders: # Guidance at approx 7.0+ # Best weight right around 0.8 - 1.2 # Works great with realistic models and non-realistic models # Works great with basic and elaborate prompts Sidenotes: # Seems to interact well with other loras # Render at low sizes and use /more /size /strength to upscale # This lora was trained on an Asian model and seems to work best with them so use <majic5>, <realisian>, or <chilloutmix> for best results. It WILL work on non-Asian models but the success rate seems much lower. Example Prompt: /render 1girl, <nipple-tweak>, tweaking own nipple, breast grab, [blurry, lowres, worst quality] <majic5>
pose	nipple-tweak2			No
pose	nipples-chain		porn boobs fetish poses bondage	a lora that creates renders women holding a chain in her mouth connected to nipple clamps Recommended settings for getting good renders: # Guidance at approx 7.0+ # Best weight right around 0.2 # Works well with realistic and non-realistic models # Works great with basic and elaborate prompt Sidenotes: # Seems to interact well with other loras # Render at low sizes and use /more /size /strength to upscale # Yes, this lora has a LOT of tokens...make sure you get them all in there. Example Prompt: /render 1girl, <nipples-chain:0.2>, chain in her mouth, nipple clamps, mouthclamp, [blurry, lowres, worst quality] <believe>
lora	nipples-dark		porn boobs fetish	No
effect	nipples-inv		effects porn anatomy boobs	No
lora	nipples-piercing			No
lora	nipples-poking		anatomy porn boobs	Produces nipples poking through clothing. Author recommends using a 2:3 ratio for best results (e.g. 512x768).
lora	nipples-puffy			No
inversion	noir		.negatives realistic art styles	No
effect	noise		effects lighting	For better contrast, and darker images. # Use LoRa' weight and additional triggers to control the intensity of the effect.
effect	noise-35mm		effects lighting retro realistic	An analog style lighting effect LoRA based on a checkpoint by the same name. The creator says the effect is mild, so crank this to 1.5 or 2.0 for that technicolor rosy hue. Tested with <realvis20>, it' cute though responds unpredictably to neon color requests
effect	noise-dark		effects lighting	Quick and easy way to reduce the amount of light in your renders. This LoRa works with almost everything. Start with weight 0.5 and adjust accordingly to your needs. If you want to make it even darker, add “dark theme” at the beginning of your prompts. Using negative weight will make everything brighter.
effect	noise2		effects lighting	No
effect	noisefilm		effects lighting	No
effect	noiseform		effects anime	No
effect	nopanties		hentai effects anime	The author wewewew wrote an excellent description for this LoRa: "This is a Lora trained on images where a girl is probably or obviously not wearing panties, but pussy is not visible (such as haitenai? tag on pixiv). Most models, especially anime/booru models, can be too eager to go full nsfw when trying to generate suggestive images, and if you add "pussy juice" then it' pretty much impossible to not generate exposed pussy. This Lora will drastically improve consistency when generating those borderline suggestive prompts, and also contains training for various types of ways to display a lack of panties without exposing the crotch. See list of tags below for details. npmaybe, probably no panties : used on images where the lack of panties is questionable, or at least not too obvious. "probably no panties" will be affected by "no panties", depending on the base model, while "npmaybe" is new so should have an equal, but lesser effect. You can use both, or either, test and see. npclearly, obviously no panties : used when there' no way girl' wearing underwear. The training didn't really grasp the difference very well, so most of time it won't make any difference which tag you use, but it does sometimes make a difference as shown in the previews. hidden privates : used on all images in the dataset. Very effective at hiding privates, as you could guess. I would recommend always using it. hidden crotch or partially visible crotch : one or the other used depending on if the crotch, specifically the area right next to where you would expect the pussy to be, is visible or not. Ex: raising skirt right at the edge so bottom of crotch is barely visible, or a front slit cutting off right next to pussy. hidden by leg/clothes/hair/tail/border/other : These help hiding the crotch, but the "by" part doesn't seem to do anything. Probably not enough training data for each. "convenient leg" was also added with hidden by leg. Other new tags : upthigh : when thigh is visible especially high, above groin. Usually on front view, skirt lift images. innerthigh : this one' ambiguous and I'm not sure it does much, but it' added on images where you can see the inner thigh right next to the crotch. sidethigh : when you can see the bare thigh from the side all the way up to the butt. Probably won't do much if the view isn't from the side, use "from side", or "side view" for more moderate effect. sitbare : when sitting down, usually on floor, with knees up, and you can see the bare bottom of butt. I haven't tested this much, I wouldn't expect a noticeable difference since default generation is like that already. backcrack : the opposite of groin I guess, lowleg from behind showing butt crack. underbutt : not a new tag, but putting it here since it' similar to above tags. underboob for butts. (slightly) lifting one side/both sides of skirt, very lowleg Relevant existing booru tags: groin, skirt slit, side slit, front slit, skirt lift, lifted by self, wind lift, skirt tug, microskirt, bottomless, naked shirt/apron/tabard/etc, no bra, convenient censoring, hair censor (check out hair censor loras, very few images with hair censoring in this one so it probably won't work great on its own), see-through silhouette (lora for this will also help a lot, otherwise it usually becomes completely see-through, although it does always hide crotch)"
style	nouveau		paintings art style	Nouveau Art style is a classic illustrated style by model maker Machine Delusions. This Lora does a lot of geometric designs and is biased towards flat vector illustration. Try it with semi-real bases like Protogen
clothing	nurse-uniform		clothes	No
lora	nwsj		female asian characters	A model that claims to be the face of the creator, it' Nauska / Nawusijia / Nausjja realistic version. Suitable for realistic base models and semi-real for fantasy/mecha/sci-fi use. A weight of 0.8 or less is recommended. Works well with other loras like <fashiongirl>
lora	object-insertion			No
style	occidental	No	.base models illustrations comics cartoons anime	This full base model is designed to create an illustrated artistic style with a western look. It is a blend of several models created by SpicyTree on Civitai.
styles	occidental		.base models illustrations comics cartoons anime	Occidental Mix by SpicyTree writes "This is an attempt to create a more "western" character look, as I've grown a little tired of seeing the same anime faces. Still very much a slightly NSFW-prone female focused base model. Does not require complex prompting - your mileage may vary prompting like you would for a more traditional anime model. Try building prompts from the ground up on this one. I've found its fairly compatible with most (non-style) LoRAs. V2 should be an across-the-board improvement, with greater coherency than V1, while still retaining the same look. Hope you enjoy the model, and I would love to see what you are able to create with it! All images generated using HiRes Fix with a denoising strength of 0.35, a scale factor of 2, and 4x UltraSharp as the upscaler. A CLIP skip of 2 was used for all V2 gens Would strongly recommend EasyNegative and (worst quality:1.4), (low quality:1.4) in the negative prompt."
style	occidental3	No		No
anime	octopus		realistic anime hentai creatures fetish memes	Octopus LoRA model, by Psoft. It can be used for a wide range of purposes, from realistic to fantasy, with particular extra training towards hentai tentacle NSFW images. Tested with realistic and anime models, it works well on both at default weight. Try it in combination with these useful tags: animal, tentacles, cephalopod eyes
style	odanon-style		anime art style illustrations	?Art Style?Oda Non Style v1.5 by zakp Recommended settings for giving your renders this particular artstyle: # Weight ~0.15-0.3 # Base model <anylora> # /parser:new /clipskip:2 Sidenotes: # Can easily handle rendering in resolutions such as 832x1024
anime	ogiwara-sayu		characters female anime	Ogiwaru Sayu - Higehiro v1.0 by Joshyy # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low # Mangled limbs, general chaos ? Some LoRas can’t handle rendering with resolutions bigger than 512x512 // Also make sure your prompt isn’t too complicated for this LoRa # No visible effect ? Check if the LoRa has any triggers that need to be added in order to activate it # Good full models to try out LoRas with: anime ? <anylora> <counterfeit> <breakanime> comics ? <comics-vision> <icomix> <misstoon> semi-real ? <revani> <lyriel16> <dreamshaper6> realistic ? <majic> <bra5> <chilloutmix> <edge>
style	oilpainting		semi-real paintings art style	No
style	old-engravings		effects illustrations art style	No
inversion	old-fashioned		art styles illustrations retro	This textual embedding is intended to generate renders made to look like pulp artwork of the 40s and 50s.
style	old-newspaper		art style retro	This lora creates front pages of late 19th-early 20th century newspapers Recommended settings for getting good renders: # Guidance at approx 7.0+ # Best weight around 1.0 # Works well non-realistic and realistic models # works well with basic and elaborate prompts Sidenotes: # Seems to interact well with other loras # Render at low sizes and use /more /size /strength to upscale. # the more elaborate your description of what you want to see on the front page, the better the render. Be creative! Example Prompt: /render a disastrous fire on the city docks, npzw, sepia, grayscale, <old-newspaper> [(worst quality, low quality:1.5), (deformed, distorted, disfigured:1.3)] <dreamshaper7>
lora	olivia-casta		characters porn female	Olivia Casta is an American-Spanish model, social media influencer, and Instagram star. She is also an adult content creator who shares risqué and explicit content on the platforms such as Only Fans and Fansly
lora	olivia-dunne		characters female realistic	A realistic likeness LoRA of gymnast and social media icon Olivia Dunne, created by Riv. It is intended for use with base models like Realistic Vision, however also works in paint models, too
pose	on-off		porn poses realistic anime semi-real effects	On/Off LoRA creates side-by-side pictures of one or multiple subjects with their clothes on and then with their clothes off, in a single image. This model is highly sensitive so please note the settings: Use a very low weight, between -0.5 to 0.5. For anime start at 0.1. Only certain base models work: <aohard> <deliberate> <urpm> <liberty> <clarity> but not others. For example, urpm 1.3 will generate a solid green wall. The overall effect is hit or miss and takes patience, not recommended for beginners
lora	on-off-boobs		boobs porn fetish	No
effect	on-off-cum		effects porn expressions	No
style	one-punch		anime art style illustrations	No
lora	onepiece-nami		characters female asian	No
pose	ontummy-reverse		porn poses	This lora renders images of women lying on their stomachs facing away from viewer Recommended settings for getting good renders: # Guidance at approx 7.0+ # Best weight around 0.8 - 1.0 # Works great with realistic and non-realistic models # works well with basic and elaborate prompts Sidenotes: # Seems to interact well with other loras # Render at low sizes and use /more /size /strength to upscale # Lora does a good job but you will definitely need some inpainting for feet and privates Example Prompt: /render 1girl, solo, ass, lying, on stomach, perfect pussy, spread legs, perfect anus, spread legs, legs up, feet up <ontummy-reverse:0.8> [worst quality, lowres, blurry]<photon>
style	openjourney		semi-real .base models environments art style	No
fantasy	orcs		creatures anime hentai male fantasy gaming	Orcs LoRa, and due to reasons, an elf one as well. A very well-made LoRa, it’s easy to use with most anime models and, to a smaller extent, with more realistic ones. In addition to putting your favorite character in very NSFW troubles, it can also generate detailed SFW orcs. Good for fantasy battle scenes. How to use: - Weight 1.0 with your usual guidance - A trigger word is not needed, but you can use the tag “orczor” for an immediate and strong effect. - Other prompts to modify or amplify the effect are listed in triggers.
clothing	oversized-shirt		objects clothes	Simple LoRa with a singular objective, but this is also what makes it work so well. It can be used both with real and anime base models and doesn't require complicated prompts. Use the prompt 'off shoulder' to trigger the falling off effect. Works well with weight between 0.5-1.
clothing	oversized-tops		clothes poses	Oversized Hoodies, Sweaters, open shoulder turtlenecks, deep v-necks, and other baggy tops. This one also has a modifier trigger for a back pose as well. It' tagged as NSFW because the top is all she' wearing, pants optional. /sampler:dpm2m /guidance: 5 thru 9 LORA weight for txt2img: reccomended 0.2-0.7 (i used 0.5 in my images). Keys: oversized_hoodie, oversized_sweater This LoRA is very easy to use, usually works with any color, just repeat the keyword with the color, like "oversized_sweater, red oversized_sweater" Try oversized_sweater or hoodie placed the keyword at the beginning of the prompt. Most of the images of the training are with hoodie/sweater with naked legs, but some of them have boots or stocking, so if you want them you might try to add it to your prompt. Use <chilloutmix> or similar as your base
pose	pain-expression		fetish expressions poses	a Lora that produces a painful expression, including tears Recommended settings for getting good renders: # Guidance at approx 5.0 - 7.0 # Weight is best right at 0.0375 but it can really vary and will be up to the user # Works great with realistic models and non-realistic models # Works great with basic and elaborate prompts Sidenotes: # Seems to interact well with other loras # A good resolution to render at is /size:640x640 & /size:640x832 # Going overboards on weighting this lora is amamzing for producing horror-themed faces Example Prompt: /render portrait photo of 1girl, a broken expression, crying, in pain, <pain-expression:0.0375> [blurry, lowres, worst quality] <realvis20>
lora	paizuri-irrumatio	No		No
lora	pale-nipples		boobs porn anatomy	No
lora	pam-beesly		female characters	a lora that creates renders of everyone' favorite Dunder Mifflin receptionist Recommended settings for getting good renders: # Guidance at approx 7.0+ # Best weight right around 0.8-1.0 # Works best with realistic models but can have interesting results with non-realistic # Works great with basic and elaborate prompt Sidenotes: # Seems to interact well with other loras # Render at low sizes and use /more /size /strength to upscale Example Prompt: /render /sampler:dpm2m /guidance:5 /size:512x768 <pam-beesly:0.8>(pambeesly:1), (realistic green eyes\|realistic hazel eyes), under eye coloration, cute, (sharp focus:1), (cinematic lighting:0.4),(realistic:0.8), low key, natural indentations in face and body, (expression lines:0.8), slight hair frizz, natural eyebrows, subtle muscle tone, [childlike, exaggerated features, blurry. lowres, worst quality> <believe>
clothing	panties-crotchless		porn hentai clothes	No
clothing	panties-toside		porn clothes	# Use with weight about 0.7-0.9 # If the lora isn't producing desired results, try this prompt: pnts, front photo of beautiful blonde woman with panties to side spreading legs, masterpiece, perfect lighting, intricate, incredible detail, perfect face, bokeh, masterpiece, good quality, [lowres, blurry, worst quality] Then you can modify it to your liking.
clothing	panties-toside2		porn hentai clothes	If you’re getting distorted renders, lower the weight. <panties-pulledaside:0.01> is a good starting point to make it work. Yes 0.01
pose	pantiesoops		poses porn anime	No
pose	pantpulldown		porn poses butts	No
pose	pantpulldown2		porn poses	No
clothing	pantyhose		fetish feet clothes	Pantyhose 20D stands for 20 DEN black pantyhose, a popular hosiery denier with medium sheerness and darker. It means those fibers have a weight of 20 grams. Denier refers to how see-through they are, whereas these are low DEN, giving legs a sophisticated look. Try a weight of 0.7 with a base like <chilloutmix>. There is also a different loRA for shiny pantyhose you can try instead.
clothing	pantyhose-shiny		clothes	A high shine pantyhose clothing model by creator Break_T. Recommended weight: 0.4~0.7, the higher the weight, the shinier the pantyhose. Works with a base of <chilloutmix> or similar
clothing	pantylines		clothes fetish butts	How to use: # If you’re getting distorted renders, lower the weight. <pantylines:0.01> is a good starting point to make it work. Yes 0.01. # In case of a ‘memory exceeded’ message, lower the number of steps, decrease the number of images & size, or try it with another base model.
style	papercut2		semi-real .base models art style graphic design	No
style	paragon		semi-real realistic fantasy paintings .base models	Paragon is a full base model that works well with many LoRAs. It' balanced between realism and fantasy with a vibe of old-school anime. It is by the creator who brought us <realvis20> Try the new recipe too: #paragon
style	peanutz		art style cartoons effects comics	No
pose	pee		porn fetish female realistic anime semi-real poses	Peeing Woman 1.2 is a squatting woman taking a leak, and boy can you do a lot more with this one. There' a handful of trigger options and the creator notes that clothed and nude both work, but clothed should add the negative for [underwear]. It works most reliably on realistic models like <urpm13> but fantasy and anime base models also work, so your devils can take a fiery leak and shoot other liquids. It was surprisingly easy to use, even at weight 1. Recommended is 0.3 weight
lora	peeking-out	No		No
object	pendulum		anime semi-real objects hentai	This LoRa is hard to use. Best results were achieved with resolution 640x640, but the coin position and number is random. A good model to start experimenting with it is <aohard>. Try with weight 0.5-0.9.
inversion	penelope-cruz		characters female realistic semi-real anime latin	Penelope Cruz Young Likeness by HazmatA17 is a postitive textual inversion. Use it just like a LoRA with a realistic or illustration base model of your choice, its quite flexible.
pose	penis		anatomy trans porn hentai penis male	No
anime	penis-anime		anatomy penis porn anime	No
lora	penis-comparison-wip			No
lora	penis-flaccid		penis anatomy futa lgbtq male porn trans	This lora is meant to be a complimentary addon to existing penis loras. It works well with futa-sm and others. It may give smoothed out results at first but use /more or /highdef to increase the quality. Lower weights are needed to keep the image coherent.
anime	penis-furry		furry penis anime porn	No
lora	penis-horse		porn fetish penis anatomy	No
lora	penis-horse2		anatomy penis porn fetish trans	No
pose	penis-licking-pov		porn poses oral	This lora is pretty self-explanatory. It produces images of women licking a penis. The ideal weight for this lora is 0.01 It was trained on, and seems to work best with, the <realflex> concept, however it works well with other realistic AND un-realistic models. It also seems to play well with other loras. Example prompt: /render 1girl, pov photo of the LickingPenisQuiron pose <penis-licking-pov:0.01> , beautiful blonde <realflex>
pose	penis-massive		penis anatomy porn memes realistic	It can create a very large realistic flaccid penis, but the author states it' more versatile than this. Try it out with a weight of 0.8.
pose	penis-on-head		poses porn penis anime	No
lora	penis-real		anatomy trans porn hentai penis	No
pose	penis-shadow		hentai porn memes penis	One of the rare LoRas that need to have its weight raised in order to work properly. Try it out with a weight ~1.5-1.7
lora	penis-shock		penis expressions porn	No
pose	penis-surrounded		penis poses porn	How to use: # It has problems with tendering in sizes bigger than 512x512. # It works best with anime based models, try it out with <anylora>.
lora	penis-uncut		anatomy penis	A natural penis lora. The creator RickMashups says pussys are common as are missing assholes, so these things are accounted for in the unusual included token
lora	people-count		detailers	This lora is designed to adjust the number of people in a render.
lora	perfect-pussy		porn anatomy	No
style	perfectworld		semi-real .base models fantasy hentai	The model maker is trying to achieve the ideal semi-realistic hentail female figure. A decadent, high quality NSFW model that sports incredible depth of field and semi-realism, makes brilliant characters
style	perfectworld4		porn semi-real paintings fantasy boobs .base models	Perfect World Version 4 is an incremental jump in quality from Version 3, which already sported impressive visuals. As described by creator Bloodsuga: "Even better faces than v3 and perhaps NSFW capabilities too, thanks to 50% of v4 being ChikMix v2 ;) Face changing LoRA/TI might be redundant for v4. Model tends to draw nsfw/females anyways so there' no need to try too hard for that. Face/style changes depending on prompts, feel free to experiment. Model has some major side effects such as freckles/bad anatomy/hands, hard to draw dark skin/squatting, and other unknown effects. You may use v2 or v1 for a more stable experience. Although not shown in examples, model can do males, scenery, etc. quite well too. Use "freckles/mole" tag(s) in negative to reduce the dots on face/body side effect if needed. The pursuit of perfect balance between realism and anime, a semi-realistic model aimed to achieve beautiful realistic faces with sexy hentai bodies. This model aims to be flexible with all kinds of NSFW/hentai concepts, so a large portion of the merge will be hentai models."
celebrity	philip-hoffman		characters realistic male	No
lora	photo-comparison			No
base	photon		.base models realistic	No
style	photoreal2		realistic .base models realistic semi-real interiors architecture	No
inpainting	photoreal2-inpainting	No		No
pose	piledrive-sex			No
pose	pillory		fetish bondage objects	This LoRa works well with most anime models but doesn’t handle real ones well. Use it with a weight of about 0.1-0.25. To get the best results, keep the rendering size at 640x640. The pillory fixed on the ground and the pillory detached cannot be specified separately now. Using a leaning forward prompt and a wide picture size is more likely to generate a pillory fixed on the ground.
lora	pillory2-realistic			No
clothing	pink-fashion		clothes	No
pose	piperperri-meme		memes porn hentai poses	No
style	pirate-punk		effects art style	This lora is designed to add a pirate theme to anything it touches. Will add skull, crossbones, and even pirate ships to nearly anything. Works well with default weights and works well with revani.
style	pixar		semi-real .base models cartoons art style	An updated that improves on the <modern-disney> model in just about every way, this so-called Pixar base model creates wonderful 3d characters in the style of the studio though don't expect it to know every franchise or character. Great for beginners, doesn't need complicated tagging. That said, here' some example keywords that might help you nail the look: character disney 3d digital art 3d render pixar style art style woman artstyle cartoon digital illustration faces male man portraits toon, octane render, unreal engine, 8k
style	pixar-a		semi-real .base models cartoons art style	Disney Pixar Cartoon Type A v1.0 by PromptSharingSam6319 <pixar-a> is a versatile model that can do both disney-like and pixar-like renders. It' more than capable for picking up the work of old <pixar> and <modern-disney> models. Tip: If your results come out blurry and unfocused, use /highdef or /more /size:1200x1200 /images:3 /strength:0.4 /guidance:10 This will greatly improve their quality.
style	pixar-b		semi-real .base models cartoons art style	Disney Pixar Cartoon Type B v1.0 by PromptSharingSam6319 <pixar-b> is a less anime looking version of <pixar-a>, it' also better at rendering old people and men. Tip: If your results come out blurry and unfocused, use /highdef or /more /size:1200x1200 /images:3 /strength:0.4 /guidance:10 This will greatly improve their quality. HOWEVER! Don't use it on inhuman looking character as they will most likely mutate.
style	pixel-portrait		gaming art style effects	Pixel Portrait by NeriJS creates lovely 16-bit style characters that look straight out of a Super Nintendo role-playing game adventure. Tested with <dreamshaper> as the art base, works best with the /tall aspect ratio
effect	pixelbringer		gaming effects retro	LoRA to make pixel art and give your renders an old-school game vibe. It already has the token ‘pixel art,’ so avoid putting it into your prompts. Treat the weight (from 0.1 to 1) as a value controlling the ‘pixelization’ amount. At full weight, <pxelbringer> is good at creating 2d sprites. This LoRa works best with anime or semi-real base models. The author recommended using it with <aohard> or <cetus>.
lora	poison-final-fight	No		No
anime	pokemon-charmander		characters anime creatures semi-real pokemon nintendo	Charmander is a monster inspired by the look of the Pokemon character by the same name. It' possible to get a realistic and anime look, depending on the base model you choose. However, it is challenging to get the realistic version as most base models will send your image into a cartoon when pokemon is mentioned. Not recommended for beginners, but it works
anime	pokemon-cynthia		nintendo pokemon characters female gaming anime	Pokemon - Cynthia / Shirona v3.0 by ARandomModelMaker # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low The training set contained multiple variations of Cynthia' signature look so you might need to put combinations of the following tags in your positive or negative prompts to get the desired results. Some should probably be used together (separated by plus in this list): fur collar, fur-trimmed coat, hair ornament, hair over one eye + one eye covered, high heels, pants, shirt + v-neck As the training set contained a few images with floating hair, you might need to add floating hair to the negative prompts if not desired. Depending on what you want to generate, you may also want to experiment with adding cleavage to the negative prompt or specifying a breast size as they vary quite a bit. Known Limitations / Problems: The shoes were not present in a lot of images and not a focus for the training so they sadly will not be consistent at all. The gems on the coat also sadly seem to be quite inconsistent despite being in a lot of images. This model seems to be much more fixated on the simple backgrounds used during training than others I've done (despite some variation in the training set), so you might need to add simple background or similar to your negative prompts. Even then, they always seem to appear from time to time. It is also sometimes a bit fixated on the 2-3 images in the training set where a covered navel was visible under the shirt so try covered navel in your negative prompts. Despite a lot of attempts, the separation between what is part of the coat vs. what is part of the shirt below also does not seem to have stuck properly yet. The coat + shirt seem to get combined to a (suit) jacket fairly often so try jacket or suit in your negative prompts if that happens. From behind, the pants tend to come out as jeans, so try jeans in your negative prompts there. From behind if not prompting for the coat, the shirt sleeves tend to have markings where the fur trims would usually be. I have no idea how to fix this yet. Maybe try lapels in the negative prompt? The hair can be wildly inconsistent, especially from behind (but that' also the case for every illustration of Cynthia I know of).
anime	pokemon-hilda		nintendo pokemon gaming characters female anime	No
anime	pokemon-marnie		nintendo pokemon characters female gaming anime	Marnie (Pokemon S/S) v1.0 by ekune # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low # Mangled limbs, general chaos ? Some LoRas can’t handle rendering with resolutions bigger than 512x512 // Also make sure your prompt isn’t too complicated for this LoRa # No visible effect ? Check if the LoRa has any triggers that need to be added in order to activate it # Good full models to try out LoRas with: anime ? <anylora> <counterfeit> <breakanime> comics ? <comics-vision> <icomix> <misstoon> semi-real ? <revani> <lyriel16> <dreamshaper6> realistic ? <majic> <bra5> <chilloutmix> <edge>
anime	pokemon-may		nintendo pokemon characters female gaming anime	Pokemon - May / Haruka v5.0 by ARandomModelMaker # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low # Mangled limbs, general chaos ? Some LoRas can’t handle rendering with resolutions bigger than 512x512 // Also make sure your prompt isn’t too complicated for this LoRa # No visible effect ? Check if the LoRa has any triggers that need to be added in order to activate it # Good full models to try out LoRas with: anime ? <anylora> <counterfeit> <breakanime> comics ? <comics-vision> <icomix> <misstoon> semi-real ? <revani> <lyriel16> <dreamshaper6> realistic ? <majic> <bra5> <chilloutmix> <edge>
anime	pokemon-misty		nintendo pokemon characters female gaming anime	This Lora is sponsored by Rule34, as it is a NSFW realistic and anime version of the female Pokemon character, misty. For semi-real (pictured), use it with a base like <perfectworld> or to get more of an authentic Pokemon look, try <aom3> with a weight of 0.8
anime	pokemon-pikachu		characters anime creatures semi-real pokemon nintendo	No
lora	pokemon-sabrina		nintendo pokemon gaming characters female	No
anime	pokemon-team-magma		pokemon gaming anime characters female	No
lora	pokimane		characters realistic female white	No
style	poodabeep		.base models cartoons comics illustrations	Pooda Beep Mix is an Illustration base model by DuskfallCrew, a comics model with a funny name (a polite way of saying PolterBitch). Check the Civitai link for prompt ideas
effect	portraits-abstract		effects paintings illustrations	Abstract portraits with a mild emphasis on black backgrounds. Mention the colors you want to see with the included triggers, like Splatter. It' designed for anime and semi-real models. Troubleshooting: Sometimes produces colors you don't want, like blue skin. To fix this, add the colors to the [negative prompt]
sci-fi	postapocalypse		post-apocalyptic sci-fi armor	This lora is designed to add post apocalyptic elements to your image. It works quite well at standard 0.75 weight. This lora was derived from a larger post apocalyptic base mod trained on post apocalyptic cosplay. It does not appear to be biased towards any one source material. Great for fallout or general post apocalyptic renders.
lora	pov-69	No		No
lora	pov-blowjob	No		No
inversion	pov-bound-wrists-missionary		poses porn hentai bdsm	No
lora	pov-bulge	No		No
pose	pov-fellatio		porn oral penis poses	No
lora	pov-missionary-anime	No		No
pose	pov-penis-on-face		anime porn hentai penis poses oral	Not to be confused with Penis on Head or Cock on Head, this is the POV version by Emberbyte. Look at the many colorful trigger options you can stack for different penis positions, or make it a group affair. Some users have commented that it is hard to control, no pun intended, so lower the weight if the dick grows out of her forehead
lora	pov-reverse-fellatio	No		No
lora	powerarmorx-01		post-apocalyptic gaming armor	X-01 Enclave Power Armor from Fallout series by Bethesda Game Studios (Mainly based on FO4). Works better with lower CFG vaules (3-6). Recommended weight range is around 0.4-0.7 depending on the model and CFG vaule. Weight could be reduced when using SD upscale. Model should know 'with helmet on', 'without helmet on' (buggy, will require further improvements) and some colors like 'black', 'ilver' and 'blue' and shiny promt. Example promts; full body shot of a person wearing black x01powerarmor in a forest / a man with a beard wearing silver x01powerarmor without a helmet on, standing next to a bridge / a woman wearing blue x01powerarmor without a helmet on next to a ruined city Extra Info: Dataset for the model was around 40 pictures including in game screenshots from my heavily modded FO4 and some pictures from the web and manually captioning images. Released as an alpha, i will try to improve it further on my free time. Zovya' RPG v3 was used for the pictures.
pose	present-ass		butts porn poses	Present Ass by WaifuWizardWong is a porn/pinup NSFW pose where, as the name implies, the character is pointing their butt at the camera. Works with illustrated models as well as realistic
effect	prettyboy		characters effects anatomy white	No
anime	princess-daisy		anime characters gaming super mario anime characters super mario gaming	LoRA trained on princess daisy from super mario. Use tokens:princess-daisy to trigger. Works for anime and photo-realistic bases. Strength recommend from 0.4 - 1, but feel free to experiment
anime	princess-kingboo		anime characters gaming super mario anime characters super mario gaming monster girls	LoRA trained on waifu princess king boo from super mario. Use tokens:princess-kingboo to trigger. Works for anime and photo-realistic bases. Strength recommended from 0.4 - 1, but feel free to experiment
anime	princess-peach		anime characters gaming super mario anime characters super mario gaming	LoRA trained on princess peach from super mario. Use tokens:princess-peach to trigger. Works for anime and photo-realistic bases. Strength best used from 0.4 - 1, but feel free to experiment
anime	princess-yui		characters female anime	No
object	product-design		graphic design	No
object	product-elegant		graphic design	No
object	product-minimal		graphic design	No
sci-fi	product-tech		graphic design sci-fi	No
pose	pronebone		poses porn	No
lora	proportions		anatomy	No
sci-fi	protectron		sci-fi post-apocalyptic mecha gaming	This lora was designed and trained by community member Eggs-Benadryl. This lora is designed to create protectron robots from the Fallout series, using AI art generation. It can produce realistic models of protectrons with various features and colors, depending on the input. The optimal weight for this lora is between 1 and 1.5, as lower or higher values may result in distorted or incomplete models. One of the challenges of this lora is to generate the grabber claws of the protectrons, which are often disfigured, just as normal fingers in ai art. . Another possible issue is that the lora may generate model stands or bases for the protectrons, which are not part of the original design. To prevent this, it is advised to use negative prompts that exclude stands or bases from the output.
style	proto53-real		realistic .base models semi-real	Protogen 5.3 Photo is the most realistic of the Protogen series, as it contains the least anime. By today' standards, it' hard to recommend for realistic pictures, but it works well with Loras and Unreal Engine style semi-reality
style	proto58-scifi		semi-real .base models fantasy	Protogen 5.8 is the so-called "sci-fi" release, because it was blended with a model called Robo Diffusion and HasDX. It' fun for semi-real and fantasy as a base model, but do note it was released in 2022
style	protonova		semi-real .base models fantasy	Protogen Nova is release 8.0 of the semi-realistic model series. Due to the confusing naming convention, its hard to keep these straight, but we'll try. Nova is most heavily weighted by Seek Mega Art, ModelShoot, and Dreamlike Photoreal2, with a little porn and anime. It' a good general fantasy model
fantasy	protoss		art style fantasy sci-fi	ProtossAI v1 by LDWorksDervlex This model creates images in the style of the Protoss folk of StarCraft 2. It is very versatile and can give you a Protoss style mech, armor, vehicles, architecture; and yes, even a Protoss coffee-maker. In the gallery page you can also see some other characters, monsters, suits, and robots. Check it out...
lora	pubes		anatomy hentai porn	No
effect	public-photo		effects realistic	LoRA trained to generate a person in public surrounded by people, such as photographers. Works with other LoRas and various anime and semi-real models. Try it out with <aohard>, <counterfeit>, <revani> and <lyriel>. If you want more people in the background, use a wider size, like 832x640, or reverse it for better focus on the character. Keep weight at around 0.8-0.9. The main trigger word ‘public’ is already included as a token. Other helpful prompts are listed in triggers. Example: /render /size:832x640 <public-photo:0.9> 1girl, photographer, denim shorts, crop top, full body, foreground, camera, cosplay, crowds, ((masterpiece, best quality, highly detailed, intricate, empty hands)), [[lowres, worst_quality, blurry, monochrome]] <lyriel>
lora	public-restroom	No		No
pose	public-toilet		fetish hentai poses semi-real porn	No
clothing	puff-dress		clothes	No
style	puffy		.base models anime semi-real hentai	Puffy v1.0 anime by Hakim3i 2.5D Anime model based on AOM3 and Sakimichan style. It' good for puffy looking character with mature bodies, and it' also very, very horny. Positive Prompts: (masterpiece), (best quality), HDR, intricate detail For girls: 1girl, solo focus, skindentation, perfect face, puffy face, beautiful face, big eyes, puffy eyes, perfect eyes, eyelashes, skin color, eye color, hair Or 1girl, solo focus, skindentation, perfect face, puffy face, beautiful face, big eyes, puffy eyes, perfect eyes, eyelashes, popular character Use body parts for full body shot example: chubby body, narrow waist, big pelvis, wide hips For Negative Prompts: (worst quality, low quality:1.4), low contrast, monochrome, text, watermark, thunder, skin imperfection Use nsfw, naked, nude, nipples in the negative if you want clothes on your girls. Recommended settings: Sampler: dpm2m CFG scale: 7 Size: 512x768 or 512x512 Clip skip: 2 Known issues: Third-party hands. Faraway eyes are not good. Use /more carefully as it can easily ruin the characteristic look.
lora	pulled-by-self	No		No
style	pulp-50s		art style illustrations retro	This art style was based on vintage art illustrations of posters and advertisements from the 1950s. CFG Scale: ~5.5-7 LoRA weight: ~0.8-1 Checkpoints: I still think Deliberate is the best choice since I tested all of my images on this checkpoint and it gives me the most accurate images to my prompts as much as possible. Denoising strength (if you're doing hires. fix): ~0.55-0.7 I generate the most accurate results with prompts that start with this: illustration, vintage art Negative prompt that works well: deformed, bad anatomy, disfigured, poorly drawn face, mutation, mutated, extra limb, ugly, disgusting, poorly drawn hands, missing limb, floating limbs, disconnected limbs, malformed hands, blurry, ((((mutated hands and fingers)))), duplicate images, duplicate, (deformed, distorted, disfigured:1.3), poorly drawn, bad anatomy, wrong anatomy, extra limb, missing limb, floating limbs, (mutated hands and fingers:1.4), disconnected limbs, mutation, mutated, ugly, disgusting, blurry, amputation,
anime	punchline-dc		characters female anime comics	No
inversion	pure-eros-face	No		No
inversion	pure-eros-face		.negatives	Pure Eros Face by IenML is a positive inversion. Add it in the positive prompt to support the idealistic 'ulzzang face' or korean idol face, aka the ?? meme. Tips from the creator: Pair it with the base model <lofi> and semi-realistic models also work, but avoid using it with 2d anime, or you'll get manly faces. When drawing high-quality faces, do not use the "detail face" tag, otherwise it may lead to deviation from the original semantics of embedding. If your generated faces often close their eyes, adding "shiny eyes, looking at viewer" will fix it.
anime	pure-furry		anime furry male	Easy to use, lora that immediately turns most characters into furry. Suitable for <aom3>, <yiffymix2>, <counter25>, and other anime models. It has a big impact even when used with low weight.
lora	pussy		anatomy pussy porn	No
effect	pussy-grool		porn effects detailers pussy	No
lora	pussy2		anatomy pussy porn	For txt2img: LoRA weight of <=0.5 You can generate the composition in txt2img, then edit further with img2img/inpainting with this LoRA HiRes Upscale: /more /size:786x786 etc For img2img: LoRA weight of 0.4-0.6 Inpainting: LoRA weight of 0.4-0.6 Masked Content = Original, works better than Fill if base image has an ok vagina already there Inpaint area: Whole Picture Some comments noted that this LoRA causes the whole image to lose resolution
lora	pussy3		anatomy pussy porn	No
lora	pussy4		anatomy pussy porn	No
lora	pussyfat		anatomy pussy porn	No
lora	pussyinnie		anatomy pussy porn	No
pose	pussyspread		anatomy pussy hentai poses	No
pose	quick-hands		anime poses anatomy	Close-up portrait hand gestures for anime portraits, a Lora that supposedly reduces hand distortion. Your mileage may vary as this is a new Lora with few reviews at the time of this update. This lora model is focused for anime base models
lora	quickfix-modesty			No
lora	rachelkay		characters female	This lora specializes in creating the likeness of twitch streamer and influencer Rachel Kay. It is intended to produce realistic and diverse portraits of her. The lora focuses on her visual appearance. Works best with the <photon> model.
effect	rage-mode		expressions effects anime	No
style	rainbow		semi-real .base models art style	Rainbow is the PG name for the Midjourney-trained model called ColorJizz, modeled with permission from creator Destiny K Rainbow on Twitter who is known for their rainbow-heavy colorful works
lora	ralph-fiennes		characters realistic male white	A model meant to resemble English actor and director Raph Nathaniel Fiennes, who gained worldwide fame for playing Lord Voldemort in Harry Potter. This LoRA is focused on his more serious film roles, and looks younger overall. Not much is known about the recommended base, but it works well with Deliberate
style	ranma		anime .base models art style	A base model focused on Anime of the 80s, with special emphasis on the Ranma series. Specific character names do work, but its also good for generating new characters. Alternatively, try the general <anime80s> lora on top of base models like this for more variety
style	ratatatat74		art style anime illustrations	ratatatat74 Style LoRA by Lykon
style	realcartoon3d		anime semi-real .base models art styles	We published this base model with former and current Daz3d users in mind. The model deliver good quality and detailed 3d-like renders with minimal instructions. Use the quality textual inversions for even shorter prompts. Unless instructed, the model is trained to draw a variety of races (African, Asian, Caucasian, Latino) across the results. Does well both male and female. Should be ok with Anime LoRas. Sweet spot: /sampler:dpm2m or /sampler:k_euler_a /size 640x832 /clipskip:2 /steps:24 and above Negative prompt for good results: (worst quality:2), (low quality:2), (normal quality:2), lowres, normal quality, ((monochrome)), ((grayscale))
style	realflex		realistic semi-real fantasy .base models porn	AgraA RealFlex is a sex-focused checkpoint merge based on both Real and Fantasy models, with special emphasis on Futa (tgirls). It' easy to prompt, and great for beginners. It can handle straight sex, LGBTQ, trans, anime, furry, you name it. The model creator Quiron recommends a /guidance between 6.5 to 9 and /sampler:dpm2m
style	realgalaxy		realistic .base models semi-real realistic interiors architecture	A high quality, very versatile base model from the makers of <galaxytime>. Note that it is semi-realistic by default so use with a strong anatomy style or negative prompt if you're going for realism. Also try "comic art" prompts with it, it can do comics! The model is officially called "ForYou". Here is some prompt inspiration to boost your results: Moody portrait, striking features, beauty, intricate details, dramatic composition, tension, wispy hair, contrast, sweaty shiny skin, texture, realism, high-quality rendering, stunning art, high quality, film grain, Fujifilm XT3, detailed skin, volumetric fog, evening, shadows
style	realisian		.base models realistic semi-real porn asian	Realisian Version 4 is a play on the words Real and Asian, but spelled without the A in Asian, so if you type it wrong 10,000 times you're not the only one. It was created by model maker Cisney Gassai. This is a full checkpoint base model blended from other models with the goal of creating a realistic model that behaves like an anime hentai model, so it' a hot one designed for porn LoRAs in particular, but also fun to try to turn your favorite anime characters into real people. Version 4 improves faces and skin.
lora	realistic-tongue-out-ahegao	No		No
style	realvis13		realistic .base models realistic	No
style	realvis20		realistic .base models semi-real realistic	Realistic Vision 2.0 is a high quality photographic model that sports some of the most believable detail, beating out stock Stable Diffusion 2.1 by a wide margin. It can also handle semi-realistic fantasy work and light illustrations. One of the better all-around starter models, it' easy enough to prompt for beginners, too. Try it with the recipe #fuji
style	realvis40		.base models realistic	Realistic Vision 4.0 by SG_161222 is an SD 1.5 based fine-tune base checkpoint, designed for realistic portraits. This update is a slight departure from previous models, as some pictures appear to have an analog film quality to them. The creator also recommends using detailer LoRAs to reduce contrast and orange tints in some cases, and has returned to suggesting 'RAW Photo' as one of the tokens, so use a recipe like #fuji that has these or type them out yourself: RAW photo, subject, (high detailed skin), 8k uhd, dslr, soft lighting, high quality, film grain, Fujifilm XT3 deformed iris, deformed pupils, semi-realistic, cgi, 3d, render, sketch, cartoon, drawing, anime), text, cropped, out of frame, worst quality, low quality, jpeg artifacts, ugly, duplicate, morbid, mutilated, extra fingers, mutated hands, poorly drawn hands, poorly drawn face, mutation, deformed, blurry, dehydrated, bad anatomy, bad proportions, extra limbs, cloned face, disfigured, gross proportions, malformed limbs, missing arms, missing legs, extra arms, extra legs, fused fingers, too many fingers, long neck
inpainting	realvis4b-inpainting	No		No
style	realvis50	No		No
base	realvis51			Realistic Vision 5.1 by SG_161222 is an iteration on their highly acclaimed realistic base model. Update 5.1 notes from the creator: "Fixes artifacts, ugly faces, masculine female faces, small eyes, poor compatibility with LoRA and TI". Example use: RAW photo, subject, 8k uhd, dslr, soft lighting, high quality, film grain, Fujifilm XT3 [[deformed iris, deformed pupils, semi-realistic, cgi, 3d, render, sketch, cartoon, drawing, anime, mutated hands and fingers, deformed, distorted, disfigured, poorly drawn, bad anatomy, wrong anatomy, extra limb, missing limb, floating limbs, disconnected limbs, mutation, mutated, ugly, disgusting, amputation <unreal-neg:-1>]]
lora	rebecca-gayheart		characters female realistic	No
base	reliberate		.base models realistic semi-real	Reliberate by Xpuct, the creator of the popular Deliberate models, surprised us with a more realistic mix of Deliberate and some other unknown ingredients. It' quite similar to Deliberate 2 in some aspects, but is designed to produce more realistic images. As this is a first generation release, it has some mixed reviews, but we love it.
lora	rena-OC-skyrim	No		No
effect	render-3d		cartoons effects	This is version 3DMM_V12 of civitai 3D rendering style lora, the more mature and less childish style of the 3D rendering series. The lora applies a 3D toon like effect to the base model, mainly big eyes and very smooth skin. Works nicely with sem-realistic 2.5 and 3D models. Good renders obtained on a weight of 0.4 to 1 on kotos, ned122, realisian, revani, paragon. No or minimal impact on toonyou that alreay has a similar body style. Know side effects: The lora blurs the background from weight 0.7 and above. You may check my test channel https://t.me/c/1638607422/103202/103579 for prompt samples to start with. Go to the channel and search for '3dmm' keyword.
effect	render-3d-alt		effects detailers	No
anime	resevil-adawong		characters gaming female resident evil capcom	A stable diffusion model for the fictional video game character, Ada Wong. Has a trigger for realistic versions, which you should also include when creating semi-real art. Use a base like <realvis13> or <revani> or <perfectworld> or <chilloutmix>. The recommended weight is between 0.7 to 1, and beyond that may overfit but will look more distinct at about 1.25, says creator s6yx
character	resevil-claire		characters female gaming capcom resident evil	No
character	resevil-ella		characters female gaming capcom resident evil	No
character	resevil-ladydis		characters gaming female resident evil capcom	No
pose	rest-stomach		feet poses	Rest on stomach, feet up (pose) v55 by rvkwi A bit finicky and hard to use, but while it works the results are good. Best way is to start with a rough description like woman lying down <res-stomach:0.6>, soles, YOURDETAILS
effect	retro-photography		retro effects	No
background	retrofuture-rooms		effects retro objects art style architecture interiors	Easy to use LoRa that works well with the simplest and more elaborate prompts. Great for making intricate room designs. If the extra details are too overwhelming, try reducing its weight to 0.5-0.8. It can also be used with negative weight to produce a much simpler, almost solitary environment. This LoRa works with most base models, but convincing it to include people and not to distort their faces takes some tinkering.
style	retrofuturism		effects retro objects art style vehicles	No
sci-fi	retrofuturism-tech		retro sci-fi effects	This lora will make your described image have a retrofuturism look with glowing glass computer screens and stylized technology. Great for fallout, sci-fi or anything involving computer systems and imagined technology. Default weight works well.
style	revani		.base models semi-real anime fantasy architecture interiors	Rev Animated v1.2.2 by s6yx ReV Animated despite its name, doesn't create animated images nor would a purist anime lover call it anime, as it is something closer to a semi-real fantasy BASE model... and of it' class, it' simply one of the best. The dynamic detailed compositions of <revani> make it the go-to favorite for many generalists. It can do a lot, and it' pleasant to work with as it generally gets anatomy right. For anime use, check out our guide: https://www.animemaker.ai/rev-animated-guide/
inpainting	revani-inpainting	No		No
clothing	reverse-bunnysuit		clothes porn hentai	Reverse Bunny Suit v1.0 by Goofy_Ai # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low # Mangled limbs, general chaos ? Some LoRas can’t handle rendering with resolutions bigger than 512x512 // Also make sure your prompt isn’t too complicated for this LoRa # No visible effect ? Check if the LoRa has any triggers that need to be added in order to activate it # Good full models to try out LoRas with: anime ? <anylora> <counterfeit> <breakanime> comics ? <comics-vision> <icomix> <misstoon> semi-real ? <revani> <lyriel16> <dreamshaper6> realistic ? <majic> <bra5> <chilloutmix> <edge>
clothing	reverse-bunnysuit2			No
pose	reverse-cowgirl		porn poses	a lora that creates renders of the reverse cowgirl pose Recommended settings for getting good renders: # Guidance at approx 7.0+ # Best weight around 0.4 # Works great with realistic models and non-realistic models # Works great with basic and elaborate prompts Sidenotes: # Seems to interact well with other loras # Render at low sizes and use /more /size /strength to upscale # Penis penetration can be adjusted using the terms "deep penetration" or "shallow penetration" Example Prompt: /render revcowgirl, naked, beautiful woman, breasts, nipples, caucasian cock deep penetration, bedroom, hands on ass <reverse-cowgirl:0.4> [blurry, lowres, worst quality] <cyberreal>
pose	reverse-cowgirl-anal-frames		poses porn	No
style	rick-morty		cartoons art style memes	Rick and Morty Characters and Style LoRA v1.0 by ez phil Trained on anything v4.5. Different characters will need different weights to work well. It' more versatile at lower weights, while at higher weights it will be more rigid but the style will be very close to the one of the show.
pose	riding-dildo		porn poses objects	Lora trained on 658 images of dildo riding in front facing and reverse facing. Can occasionally do anal just not often. Regular pose is from front, add "ass, from behind" to get rear view. Dildo shape size color features etc. are all random unless you figure out a way to control it. Trained on NAI, gen'd on AOM2 nsfw
pose	riding-sex		porn poses	This lora renders images of women in riding sex position Recommended settings for getting good renders: # Guidance at approx 7.0+ # Best weight around 0.8 - 1.0 # Works great with realistic and non-realistic models # works well with basic and elaborate prompts Sidenotes: # Seems to interact well with other loras # Render at low sizes and use /more /size /strength to upscale # Very robust and versatile lora. You can vary perspectives and angles by changing size or by using prompts such as "open mouth, looking up, looking down, from side, from above, from below" Get creative! Example Prompt: /render <riding-sex:0.9>, ridingsexscene, 1girl, closed eyes, nude, realistic, from below, mouth open, [worst quality, blurry, lowres head out of frame] <photon>
celebrity	riley-reid		characters female realistic	No
pose	rimjob		oral. porn poses fetish	LoRA trained to rimjob / anilingus sex act, where the character getting rimmed is kind of upside-down. Recommended settings for getting good renders: # Guidance at approx 7.0+ # Best weight around 1.0 and over # Works best non-realistic models # works well with basic and elaborate prompts Sidenotes: # Seems to interact well with other loras # Render at low sizes and use /more /size /strength to upscale. Works best with square aspect ratio (512x512) Example Prompt: /render <rimjob:0.9> rimjob, 1girl, penis, testicles, mature man, licking, pubic hair [multiple hands, grab, worst quality, blurry, lowres] <hassaku>
anime	rkn-chan		characters female anime memes	RKN-Chan - LoRA v2.0 by RKN-CHAN # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low # Mangled limbs, general chaos ? Some LoRas can’t handle rendering with resolutions bigger than 512x512 // Also make sure your prompt isn’t too complicated for this LoRa # No visible effect ? Check if the LoRa has any triggers that need to be added in order to activate it # Good full models to try out LoRas with: anime ? <anylora> <counterfeit> <breakanime> comics ? <comics-vision> <icomix> <misstoon> semi-real ? <revani> <lyriel16> <dreamshaper6> realistic ? <majic> <bra5> <chilloutmix> <edge> Appearance: aqua hair, long hair, sidelocks, blunt bangs, twintails, blue eyes Clothes: blue pants, blue jacket, black necktie, hair ornament, rkn symbol
celebrity	robert-pattinson		characters male white	No
lora	robot-girl	No		No
style	roha-style		anime art style illustrations hentai	No
lora	ronald-mcdonald-chan	No		No
anime	rosalina		anime characters gaming super mario anime characters super mario gaming	LoRA trained on princess rosalina from super mario. Use tokens:rosalina to trigger. Works for anime and photo-realistic bases. Strength from 0.4 - 1, but feel free to explore from 0 - 1.5
style	rpg5-beta		fantasy realistic semi-real environments .base models	No
effect	ruins-light		effects art style lighting environments environments	Good for large-scale background images with anime-models. Can try with photorealistic models
lora	running-makeup		expressions porn hentai effects	No
lora	ruriko-kojima		japanese asian characters female	No
lora	russiandoll		characters female realistic semi-real anatomy white	The RussianDoll LoRA is used for creating the likeness of non-specific beautiful Caucasian women. This V3 update works better in Realistic and Semi-Real 2.5D, however sometimes generates cleft chins. We have accounted for this in the included tokens, but may still occur when guidance is set above 7. It is also advised to use a lora weight lower than 0.4 combined with a base of <chilloutmix> for best results. The LoRA has good reviews and works well, but there are notable comments about the Russian authenticity of the look, so you may also want to add specific characteristics or other geolocations and put [asian, slavic, etc] in negative prompts when using Chilloutmix as your base. For an alternative look, also try <eudoll>
inversion	ryan-gosling		characters male realistic	Ryan Gosling by SadVideoCard, the prominent Canadian actor. This is a positive textual inversion - pair it with <realvis51> or similar
lora	saggy-pregnants	No		No
clothing	sailor-swimsuit		clothes	No
lora	salmahayek		characters realistic female latin	No
style	samdoesarts		semi-real art style anime illustrations	No
pose	sandwiched		poses anime	This concept LoRA allows you to put a character between other two other characters and have a sandwich. The weight can be increased beyond 1.0 for additional effect. The trigger word is 'andwiched'. But you also need to have something like: 2boys, multiple boys, 2girls, multiple girls You can also increase the weight of the trigger word, besides the LoRA weight.
style	sanic		anime .base models illustrations art style art style	No
anime	sao-asuna		characters anime female	Asuna from the anime Sword Art Online (SAO) is a long haired brown eyed young woman in a white and red tunic, which can also appear under plate armor. It is intended with use with anime base models like <aom3> with a weight of around 0.6. Some anime models cannot create her likeness, says model creator Lykon
lora	sao-sinon		characters female	Shino Asada is a fictional character who appears in the Sword Art Online series of light novels and anime by Reki Kawahara. She is more commonly known by her avatar name, Sinon, her player name in the Gun Gale Online and Alfheim Online video games that the series is set in
styles	sapphire		.base models anime cartoons	Mistoon_Sapphire by Inzaniak. Sapphire is a variant of my SD model Mistoon_Anime that tries to achieve a more "2.5d" style while still mantaining some resemblance to the modern anime style. The starting point for this model was my blend. From there, I then added a few of my custom LoRAs in the mix to change the esthetic. This merge has been created using a lot of different models. In the start I've used popular models (and also nieche ones), but now I'm mostly using my custom LoRAs to finetune the model in the way I like. This model is oriented towards creating girl portraits, but the V2 is also capable of creating male characters in a convincing way.
lora	satomi-ishihara		jav porn asian characters female japanese	No
effect	saturation		effects lighting	No
effect	saturation2		detailers effects	No
anime	saya-takagi		female anime characters	No
lora	scared		anatomy expressions hentai porn	No
celebrity	scarlettjohansson		characters realistic female white	No
celebrity	scarlettjohansson2		characters realistic female white	No
anime	school-rooftop		anime environments	An anime style school rooftop in Japan, a common trope in hentai and videogames. Some have noted that it may affect the character' clothes, so you may want to add a negative tag for things like sailor shirts, big breasts, and black pantyhose etc. Tested with <anylora>
sci-fi	sci-fi-bg1		backgrounds sci-fi cyberpunk 3d	Sci-fi Backgrounds EP1 by hk4life Introducing Sci-fi Backgrounds EP1 Model: Immersive Atmospheric Backgrounds Sci-fi Backgrounds EP1 Model, driven for 20 epochs and 4800 steps, is the first model of a series dedicated to creating immersive atmospheric backgrounds with a focus on sci-fi, 3D, and cyberpunk themes. This AI model excels at generating visually captivating environments that transport users to futuristic worlds, incorporating elements like advanced technology, vibrant neon lights, and cybernetic landscapes. Max weight: 0.7
celebrity	scully-xfiles		characters female sci-fi realistic white	Dana Scully from X-files by HazmatA17 is a realistic or illustrated likeness of Gillian Anderson. Use it with your favorite realistic or semi-real base art model.
character	sdv21		realistic .base models semi-real	No
effect	see-through		effects clothes anime	Makes the clothes translucent. Use triggers: shirt, dress, sundress, see-through silhouette together with the Lora. Works well with anime-related models
clothing	seethru-top		porn clothes hentai semi-real realistic	seethru-top lora applies sheer transparency to the model' clothing for an extra bit of naughtiness. Note it can be applied to both top and bottom clothing. Prompt examples are: seethru, seethru clothes, seethru classic intricate shirt, high cut white transparent leotard It will help your render to indicate which clothing part to selectively apply the desired effect.
celebrity	selenagomez		characters realistic female latin	No
pose	self-breast-sucking			No
pose	self-wedgie		anime poses	How to use: # Try it out with weight 0.5-0.9 # Has a tendency to render only the body. Describing the face may help with putting it inside the frame. # It works best with anime based models, try it out with <anylora>.
pose	self-wedgie-2		porn poses memes	No
pose	selfbuttgrab		poses porn hentai butts	No
pose	selfie		poses	No
effect	selfie-finger		porn hentai effects	How to use: # Try it out with weight ~0.75 and <anylora> # Control the effect by adding more prompts from among the triggers.
pose	selfie-mirror		porn hentai effects poses	No
pose	sex-standing-double		anime poses hentai porn	An anime double penetration pose intended for AO Hard. Some reviews noted that it is unstable, so try different weights
pose	sexafter1		poses expressions porn hentai	This LoRA enhances the "after sex" concept primarily with lying semi-senseless on a bed covered in cum. Most training data contained lying on a stomach or back, but you can also try other poses. Works with anime and semi-real base models. Try it first with size 640x640 and weight ~0.5-0.8. Prompts that may help in getting it working are listed as triggers.
effect	sexafter2		expressions porn anime hentai	How to use: # If you’re getting distorted renders, lower the weight. <sexafter:0.3> is a good starting point to make it work. # For better effect add some cum related prompts.
pose	sexafter3		poses porn effects	No
clothing	sexy-attire		porn clothes hentai semi-real realistic	Sexy Attire \| Irresistible Sexiness by prolific creator EDG from Civitai. The lora will dress your model with a lace strappy one-piece body or nuisette lingerie. Go to the author' page for example. Start with a lora weight of 0.6 to 0.75
clothing	sexy-cow-costume		clothes	No
lora	sexy-deathclaw		creatures female gaming monster girls	No
lora	sexy-firefighter		characters male realistic clothes	No
clothing	sexy-maid-dress	No		No
pose	sexy-nelson		poses butts porn anime	No
pose	sexy-nelson2		poses butts porn anime	No
lora	sexy-office-uniform		female clothes	Sexy Office Uniform / Sexy Office Lady concept by SleepBreaker is a non-specific woman' uniform that generally features a short skirt, vest, and dress shirt in an office setting. The recommended base model to pair it with is <cyberreal>
anime	sf-chun		characters anime gaming female street fighter capcom asian	Character lora focused on Chun-Li from Capcom' Street Fighter. It contains a variety of Chin-Li' versions and requires some additional prompting to bring out its full potential. <cetus> makes her look really good, but other base models also work. As proposed by the author: Good General Tags: brown hair, wide hips, thick thighs, muscular female, kicking, fighting stance, perfect brown eyes with white sclera, flexible SF2 Chun (Chun Classic): chinese clothing, blue dress, double bun, pelvic curtain, pantyhose, bun cover, puffy short sleeves, sash, white footwear, SFA Chun (Fortnite Chun): blue vest, unitard, blue pants, skintight, yellow ribbon, short hair, sneakers, bare shoulders (Gets the best look without Wide Hips and Thick Thighs tags) SF5 Chun (No-Bun Chun): chinese clothing, black dress, gold trim, pelvic curtain, ponytail, double braid, long hair, cleavage (This is the least consistent one, you may want to put Pantyhose in the negative tags) SF6 Chun (Aul Lady Chun): mature female, makeup, yellow ribbon, chinese clothing, cleavage cutout, white shirt, tight pants, blue pants (Looks more accurate without the Muscular Female tag, you may want to put Unitard in the negative tags so as to not confuse it with SFA' costume)
anime	sf-juri		characters anime gaming female street fighter capcom asian	No
anime	sf-menat		characters female gaming	No
anime	sf-rainbow-mika		characters female gaming street fighter	No
character	sghammond		characters male realistic white	No
anime	shalltear		characters female anime	Shalltear is a gray haired, red-eyed female character from an indie anime by Sergey Manokhin. The examples show options with "fang out" but isn't listed as an official trigger. Try it with strength around 0.7 with <aohard> and other anime and semi-realistic base models
anime	shalltear2		anime porn characters	A character lora focused on the anime Shalltear, this is the "bloodfallen" focused one with may triggers including a disguise, a floating pool of blood and other effects. This is the NSFW one. Use with 0.85 strength and an anime base like <aom3> or <aohard>
celebrity	shaniatwain		characters female realistic white	No
effect	sharpen		effects	Lora trained to adjust the sharpness of your image. No tokens required. The recommended weight is 0.1 - 0.4, where higher weight is sharper the image. For negative weights, works well from -0.1 - 0.5 to make the image smooth or blurry, depending on the model
style	sharpie		art style cartoons effects anime illustrations comics	This lora is designed to add thick lineweight and a more simple shading style. The effects can be beautiful and work on all animated models. Default works well but it can be increases if needed. From testing, if you use /more a lower guidance of about 6 or 7 works well to keep details fairly simple which will keep the effect noticeable and pleasing however other combinations of setting will also work.
lora	shay-laren		characters female realistic	No
lora	sherilyn-fenn		characters female realistic	A likeness of a younger actress Sherilyn Fenn during her Audrey Horne years in the television series Twin Peaks, created by model maker Garand343635. Works with realistic base models like <believe> and similar
anime	shimakaze		characters anime gaming female	No
clothing	shimakaze-cosplay		clothes	No
style	shinkawa		gaming art style retro illustrations	No
lora	shiro-tower-of-fantasy	No		No
anime	shirt-pull		anime semi-real boobs porn	A lora for generating renders where a female has her shirt being pulled down from the front. Tokens used are shirt pull. Experiment with the prompt words: motion lines, no bra, open mouth, shirt pull, bare shoulders, collarbone, unaligned breasts, large breasts, bouncing breasts, upper body, !, spoken exclaimation mark, (describe your background) background. Recommended models are to use aom3, aohard. Generally any semi-real or anime models work well.
clothing	shirt-tug		poses clothes	a lora that creates renders of women tugging on their shirt Recommended settings for getting good renders: # Guidance at approx 7.0+ # Best weight right around 1.0 # Works great with realistic and non-realistic models # Works great with basic and elaborate prompts Sidenotes: # Seems to interact well with other loras # Render at low sizes and use /more /size /strength to upscale # If you get renders without shirts or with torn shirts place extra weight on the word "shirt" in your prompt Example Prompt: /render 1girl, shirt, naked shirt, shirt tug <shirt-tug>,[blurry, lowres, worst quality]<paragon>
clothing	shorts-anime		clothes anime porn	Panties Anime (aka Modesty Shorts or Safety Panties) is a trope from anime. It also generates gym shorts. Important: Works best with <counter25> as the base (or similar) at a very low weight. Try it with a weight between 0.1 - 0.3
lora	shower			No
lora	shrek		characters male creatures memes cartoons	No
celebrity	shriya-saran		characters female realistic indian	Shriya Saran Bhatnagar also known as Shriya Saran, is an Indian actress who primarily works in Telugu, Tamil, and Hindi films. Although Saran aspired to become a well-known dancer, she became an actress and made her film debut with the Telugu film Ishtam. Saran had her first commercial success with Nuvve Nuvve
lora	silver-wolf-honkai	No		No
effect	simplenegative		detailers	SimpleNegative v1.0 by AutoRunMech Slightly enahnce image quality and hand correction. Play with it' weight to get better results, the author recommends trying from -0.5 up to ~3.
inversion	simplenegative-v2	No		Simple Negative by Aikimi is a variation on EasyNegative. It works the same way. Remember to put it inside your negative prompts with a negative weight, like this: [[<simplenegative-v2:-1>]]
style	sinner-style		art style illustrations cartoons	No
helper	skin-blemish		detailers anatomy realistic	Skin Blemish by Polyhedron AI is the 1.1 of Skin & Hands, which is a new-trained variant of the skin and hands model. Try it with a VERY LOW WEIGHT of 0.2 or 0.3 or the person will look diseased. Use the base model <photon> or <realvis40>. The creator notes that it works best with an eye detailer, and don't forget your own prompts describing the skin. Use /highdef to fix the eyes Example: RAW photo, portrait of a beautiful blonde woman wearing a red shirt (high detailed skin:1.2), 8k uhd, dslr, soft lighting, high quality, film grain, Fujifilm XT3 <skin-blemish:0.2>
lora	skindentation-wide-hips	No		No
pose	skirtlift		poses porn hentai	No
lora	slap-mark	No		No
pose	sleeping-face		poses	No
anime	slimegirl		characters anime creatures hentai monster girls	No
effect	slimegreen		effects memes	No
effect	slimewhite		effects	It makes everything covered by.., um... mayonnaise(?) Try it with a weight of 1 or higher to make white goop coat everything like the end of the GhostBusters movie. This one seems more sensitive to guidance than others, so try it with 7 or lower. It works well with models like <deliberate2> and <revani>
lora	snes-boxart		gaming	No
clothing	socks		poses clothes fetish feet	No
clothing	socks-and-sandals		clothes memes	No
pose	solescowgirl		poses hentai butts porn	No
anime	souleater-blair		characters anime female	Blair-Soul Eater v2.0 by Ciel_Phantomhive # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low # Mangled limbs, general chaos ? Some LoRas can’t handle rendering with resolutions bigger than 512x512 // Also make sure your prompt isn’t too complicated for this LoRa # No visible effect ? Check if the LoRa has any triggers that need to be added in order to activate it # Good full models to try out LoRas with: anime ? <anylora> <counterfeit> <breakanime> comics ? <comics-vision> <icomix> <misstoon> semi-real ? <revani> <lyriel16> <dreamshaper6> realistic ? <majic> <bra5> <chilloutmix> <edge>
anime	souleater-crona		characters anime	Crona (soul eater) - LoRA v1.0 by Iloveai152002 Use it with weight of about 0.4-0.6. Initial renders may look bad, but it quickly improves on quality after using on it: /more /size:640x832 /strength:0.4 /guidance:10
anime	souleater-medusa		anime characters female	Medusa Gorgon-Soul Eater v1.1 by Ciel_Phantomhive # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low # Mangled limbs, general chaos ? Some LoRas can’t handle rendering with resolutions bigger than 512x512 // Also make sure your prompt isn’t too complicated for this LoRa # No visible effect ? Check if the LoRa has any triggers that need to be added in order to activate it # Good full models to try out LoRas with: anime ? <anylora> <counterfeit> <breakanime> comics ? <comics-vision> <icomix> <misstoon> semi-real ? <revani> <lyriel16> <dreamshaper6> realistic ? <majic> <bra5> <chilloutmix> <edge>
sci-fi	spaceknights		armor cyberpunk sci-fi	SpaceKnights by Licot To create sci-fi images of the future. In order to get the necessary direction, use Trigger Words: helmet, science fiction, armor, power suit, bodysuit, cyberpunk, power armor
pose	specimen		poses effects	No
anime	spider-arachne		characters anime creatures monster girls	No
lora	spider-gwen		characters female comics cartoons	No
anime	spider-princess		characters creatures anime female	No
lora	spider-real		horror creatures	EdobSpider v1.0 by edobgames Great at making realistic looking spiders or... anime spider-girls. Mix it with other spider-girls loras at low weight for interesting results. For realistic spiders use <believe>, <paragon>, <revani> or <reliberate> For spider-girls <identity> or <monstergirls>
style	spider-verse		comics art style	A cell-shaded comic book style LoRA by Lykon inspired by Into the Spider-verse. Use it with <anylora> as the base, or similar
effect	spiral-eyes		effects expressions anime	No
pose	spitroast		poses porn oral	No
pose	spitroast2		porn poses	No
pose	spitroast3		oral porn poses hentai	How to use: # If you’re getting red noise renders, lower the weight. <spitroast3:0.1> is a good starting point to make it work. # In case of a ‘memory exceeded’ message, lower the number of steps, decrease the number of images & size, or try it with another base model. # It has problems with tendering in sizes bigger than 512x512. # It works best with anime based models, try it out with <anylora>.
pose	spooning		porn hentai poses	Spooning Position LoRA v1 by schneed This lora s a bit volatile, but inpainting the small off-details should do the trick. It is also rather prone to doing anal. Try '(anal)' in your negative prompts to counteract this. This model doesn't require any specific keywords except general sex or sex adjacent tags - provided in the triggers section. It has futa tendencies.
lora	spread-pussy-2			No
pose	squat		poses	A "Slavic" squat pose, anime or realistic, frequently seen in anime as a student upskirt trope. Use with base AV3 or Chilloutmix. Recommended weight is 0.6-0.8
lora	squatting-double-penetration	No		No
lora	squigly-skullgirls	No		No
pose	squirtomania		female sex poses	This lora renders women having squirting orgasms Recommended settings for getting good renders: # Guidance at approx 7.0+ # Best weight around 0.8 # Works well non-realistic and realistic models # works well with basic and elaborate prompts Sidenotes: # Seems to interact well with other loras # Render at low sizes and use /more /size /strength to upscale. Works best with square aspect ratio (512x512) # Adjust the weighting on "(female ejaculate from pussy:x)" where x increases the amount of ejaculate Example Prompt: /render NSFW, (full body photo:1.6), 1girl, solo, nude, (pissing in the livingroom), (female ejaculate from pussy:1.5), <squirtomania:0.8> [worst quality, lowres, blurry] <photon>
lora	sssniperwolf		characters female realistic gaming white	No
sci-fi	staite		characters female sci-fi	Lora for Actress Jewel Staite known for her roles in Firefly and Stargate Atlantis. Pretty easy and versatile run at around 1.0 weight. Can even run at 1.1 Works best with realistic but works great with semi-real and is passable in anime even.
lora	standing-kiss-handjob	No		No
lora	standing-sex-pov	No		No
lora	standing-split-sex	No		No
sci-fi	star-destroyer		sci-fi environments objects star wars spaceships	Star Destroyer (Star Wars) v1.0 by timmbobb The LORA has two activations, one without token and with it. # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low # Mangled limbs, general chaos ? Some LoRas can’t handle rendering with resolutions bigger than 512x512 // Also make sure your prompt isn’t too complicated for this LoRa # No visible effect ? Check if the LoRa has any triggers that need to be added in order to activate it # Good full models to try out LoRas with: anime ? <anylora> <counterfeit> <breakanime> comics ? <comics-vision> <icomix> <misstoon> semi-real ? <revani> <lyriel16> <dreamshaper6> realistic ? <majic> <bra5> <chilloutmix> <edge>
sci-fi	star-trek		environments star trek sci-fi spaceships	Star Trek - Ships Enterprise v1.3 by sevenof9247 To boldly go where no one has gone before! Keep the weight to no more than about 0.9. If it comes out well, copy the seed-number and slightly change the guidance, steps, strength, or sampling method. Live long and prosper.
lora	starbuck-battlestar	No		No
effect	stickers		cartoons effects illustrations	No
lora	stormi-maya		female characters black	a lora that creates the likeness of adult model/social media personality Stormi Recommended settings for getting good renders: # Guidance at approx 7.0+ # Best weight right around 1 # Works great with realistic models and non-realistic models # Works great with basic and elaborate prompts Sidenotes: # Seems to interact well with other loras # Render at low sizes and use /more /size /strength to upscale # This lora has amazing hair, add "big afro", "beautiful huge afro" etc to get some great results Example Prompt: /render 1girl, solo, beautiful woman, StormiMaya, beautiful big afro, <stormi-maya>[blurry, lowres, worst quality] <believe>
clothing	street-fashion		clothes	No
lora	strip-club		porn environments	Strip Club Version 1.2 by EDG is a multi-room environment LoRA, which you can control by the various trigger options we've documented here. Online comments noted bias towards female strippers and Asian faces, but in our tests we even got Mr. Kevin Hart to work the pole, so try different base models if you're not getting the results you're looking for.
lora	stuck-washing			No
lora	stuck-washing-anime		poses hentai memes porn objects	No
anime	succubus		hentai anime characters	No
effect	sun-shadow		effects lighting art style illustrations paintings environments	Noise-offset version of light-shadow LoRA. Recommended weight used is from 0.4 - 0.8, no tokens are required. The following prompt words can be used to enhance the image: cloud, colorful, starry, stars
effect	sunburn-tanline		effects porn	This is an extreme sunburn tanline model by TheGipper, very nsfw. It seems to only work at a very low weight like 0.2 with models like <photon> or otherwise you'll get very interesting bacon salmon alien nudists
anime	super-robot		anime cartoons armor mecha military	Super robot diffusion by waomodder The myth of steel spun by artificial intelligence. This Lora is for generating robot and mecha of various genres that might appear in an anime. It supports a wide range of designs, from heroic to military. In addition, this Lora is not intended to output any particular well-known robot. Recommended Setting Checkpoint: AnythingV5, abyssorangemix3AOM3_aom3a1b, CounterfeitV30_v30 Guidance: 10 Lora Weight: 0.6-0.8 Style Description By entering trigger words specifying various styles, it is possible to draw different designs for different genres. ?Super Robot Style (SRS) This is a robot genre with mainly fancy decorations and heroic designs. They are decorated with crystals on their chests, horns on their heads, etc. I just want a robot with a cool design! Use this one if you want a robot with a cool design. Sample prompt: ((masterpiece)), (SRS), (a ((super robot)) with sleek and menacing design:1.2), (large head:1.8),(full armor:1.5), (full body),(heroick parts), glowing eyes, full body, highly detailed, <super-robot:0.8> ?Military Robot Style (MRS) Unlike Super Robot, this robot genre has no fancy decorations and has a military weapon-like, martial and massive design. Use this if you want a muddy, down-to-earth robot. Sample prompt: (MRS) ,super fine concept art, extremely detailed, a heavy mech, hard surface, hawken, sci-fi, battlefield, <super-robot:0.8> ?Real Robot Style (RRS) This robot genre falls between super-robot and real-robot. (looks a lot like Transformers). They are mainly stylishly designed and have a slight military component. Sample prompt: ((masterpiece)), a (((mecha))) with sleek and menacing design, glowing eyes, full body, highly detailed, (RRS) <super-robot:0.8> ?Fantasy Robot Style (FRS) This robot genre has an armoured form and armour, and is decorated in a gothic style. The designs are similar to those of super robots, but are reminiscent of warriors such as knights and samurai. It is more difficult to adjust the design to both mechanical and armoured elements, so it can work well if you supplement the parts and elements with prompts. Sample prompt: ((masterpiece)), (FRS),(a ((mecha)) with sleek and menacing design:1.2), glowing eyes, full body, highly detailed, (heroic knight armor:0.8), (heroic parts:1.5) <super-robot:0.8>
lora	supergirl		characters realistic female white comics semi-real	No
lora	supergirl-tv	No		No
style	sushi25		.base models semi-real anime fantasy	Dark Sushi 2.5D v2.0 by Aitasai This model strives to bring its generated images closer to 2.5 than traditional anime, but it can do both and not only! Compared to mix version, the colors are darker and more saturated.
style	sushimix		.base models anime illustrations cartoons	Dark Sushi Mix brighter version by Aitasai This model is good for traditional anime/cartoon renders without adding any additional dimension to them. Compared to the 25 version, the colors are lighter and brighter. Its style may be compared to misstoon.
lora	susu		characters female realistic asian	No
clothing	sw-imp-officer		characters sci-fi star wars clothes	No
clothing	sw-tie-pilot		characters star wars sci-fi clothes	No
clothing	swimwear-onepiece		clothes	No
inversion	swirling-magic		effects backgrounds	No
anime	switch-chan		gaming female characters anime	No
style	sxz-horny		art style anime illustrations	It can easily produce good results, but faithfully reproducing this style will require tinkering with weights (about 0.7-0.9) and base models. The author recommends adding those prompts at the end: volumetric lighting, best quality, masterpiece, intricate details, chromatic aberration, realistic
lora	sydney-sweeney		characters female white realistic	No
style	synthwave		semi-real .base models environments art style graphic design	No
pose	tachi-e		anime effects poses	The term Tachi-e is a 19th century reference to "standing puppets", which is also now used to describe a standing full body pose in anime and games. Therefor, this LoRA that gives standing full body (tachi-e) with background. Suitable for portrait and landscape renders. Use white background, full body for tachi-e with background. Use simple background, full body for tachi-e with no background. Use white background, full body, border, outside border for tachi-e with background and border cut-out. Use white background, no human for scenery. Use Danbooru tags with a base model that allows it, like <cetus>
clothing	tacticool		clothes military	Do you like woman in plate carriers? Tattoos? If so this is for you. Adds plate carriers, gunbelts and other tacticool accessories. Most examples are of semi-real models.
lora	taf	No	asian female japanese chinese korean anatomy	Top Asian Face by IenML is a lora trained on 2000 images of beautiful Asian female faces to add more characters and variety to your renders. Try it with a base like <lofi>
lora	taiwandoll		characters realistic asian anatomy	No
lora	taiwandoll2		characters realistic asian anatomy	No
sci-fi	taldarim-mecha		anime mecha sci-fi armor	TalDarim Mecha by Cinsdia Recommended for Mecha: anylora model, steps: 20, CFG scale: 7, clip skip: 2 No triggers needed! This model makes some cool looking Mechs, but don't pigeonhole it by thinking that is all it can do, It also makes some REALLY nice fantasy armor too.
style	tanjore-art		art style indian	Tanjore Art is an Indian-styled artwork LoRA by MonarchSung. Use it with a base like <revani> and a weight of 0.8
effect	tanliner		effects porn	No
lora	tape-gag		fetish bondage porn hentai	Tape Gag v2 by 1032456 Recommended settings for getting good renders: # Keep guidance low! About 5-7 # Weight ~0.5-1.0 # Works with most anime based model and not only # /parser:new /clipskip:2 (optional) Sidenotes: # Mixes really well with character LoRas # A good resolution to render at is 640x1024
lora	tat-girls		tattoos female characters realistic	Tatgirl aka NSFW Tattooed Girls by KSOL8 is a model that generates nude women with large Western style tattoos on their arms and chest, a style popularized by the pinup publication Suicide Girls. This model emulates that style of photo without using specific tagging of actual persons. Usage notes from the creator: Types of tattoos: full body tattoo, chest tattoo, arm tattoo, thigh tattoo, leg tattoo, belly tattoo, ass tattoo (although may provide mixed results) Types of positions: sitting, standing Types of images: nsfw (use for NSFW images, strictly compulsory) or clothed (use for SFW images) Recommended LoRa strength: 0.7, you may experiment though Recommended Image size: Use portraits (512x768 etc) Tips: You can use skirtlift or shirtlift loras as this one had that kind of dataset images. Realistic Vision and Deliberate output some great results as you can see in the samples, you may also use LazyMix although I haven't tested it. Add nude, naked in negative prompts when using clothed keyword in positive prompt. Always add nsfw keyword in prompt to generate proper nude images
effect	tattoo		tattoos detailers effects	A tattoo lora by Astarote to enhance tattoo cohesion in base models that don't do them well, and perhaps to put tattoos on objects and animals that wouldn't normally have them. The creator noted that it doesn't work well when prompted with kimonos, so please watch out for that limitation. Try it with realistic models or anime concepts like <breakanime>
effect	tattoo-face		detailers tattoos effects	Face Tattoo by Nlo is a LoRA that adds facial tattoos to base models that don't have enough variety in their training data, or changes up the effect on those that have it. He noticed that even realistic models like Homoerotic have limited tattoo data, so this mini model was created to solve the problem. Limitations: Because faces are such a big part of the data, you can expect changes in the face of your subject when this is used. You may need to lower the weight of other character loras to avoid conflicts
lora	tattoo-yakuza		tattoos	Irezumi is the Japanese word for tattoo, and is used in English to describe the distinctive type of large tattoos often seen in Yakuza (gang) movies and videogames with strong red accept patterns. This loRA by BadBro works well in creating this effect but has certain limitations: The LoRA has face data, and will change the look of your character. It was trained with 640*640. It was confirmed that it works well a in 640x768 full body shot. The tattoo will wrap around certain parts of the body, or wrap around others depending on the angle. Suitable for anime, semi-real, and realistic base models
lora	tawna-crash	No		No
lora	taylor-swift		characters female realistic	Taylor Swift v1 by malcolmrey LORA of Taylor Swift - easy to use, but may require adding the prompt 'SKS WOMAN' for better resemblance.
lora	taylorlautner		characters realistic male white	No
sci-fi	tealc		characters sci-fi male black	No
sci-fi	techpunk-mask		objects sci-fi	No
lora	teenrobot-jenny		characters cartoons	Jennifer "Jenny" Wakeman (designation: Robotic Global Response Unit XJ-9) is the protagonist of My Life as a Teenage Robot. She is a state-of-the-art robotic protector of Earth, armed with advanced weaponry and devices at her disposal, but wishes to live the life of a normal teenager
anime	tekken-julia		anime characters female gaming	a Lora for the character Julia Chang from Tekken Works pretty well at 0.8-1.0 might need some fiddling with the prompt.
lora	temari-naruto	No		No
clothing	tennis-outfit		clothes	No
lora	tentacle-mouth		creatures horror expressions	If you want to make your girl special, this is a LoRa for you. Use it with low weight (0.1-0.4). Without any triggers, the results will be deceptively innocent, but as soon as you start adding them, this LoRa will show its true face - literally. Anime models work the best with this one, but you can also try others. An alternative name for this lora that ultimately was voted out was <teeth and tentacles and where' my holy water>
inversion	tentacle-ti		hentai porn fetishes creatures horror	Corneo' Tentacle Sex TI Embedding. Try it with <breakanime> or your other favorite AI model. It is a positive embedding, so use it like LoRA in the positive prompt box
lora	tentacles		hentai porn fetish creatures	Tentacles - Rindou v4.2 by psoft Recommended settings for getting good renders: # There is a bias towards the LoRa' trained style and characters. You can counter this by reducing the weight (0.3-0.7) and by adding negative prompts such as [x_hair_ornament], [bodysuit] # Try it out with base models such as <meinah3>, <cetus>, <breakanime> or <counterfeit> # /parser:new /clipskip:2 (recommended) Sidenotes: # Using too many triggers at once can turn into a traumatic experience
lora	testament-guilty-gear	No		No
lora	testicle-sucking			No
clothing	thai-pants		clothes	Fashionable Thai Pants "Kang Keng Chang" popular with tourists in Thailand. Responds well to different colors like red, purple, yellow, gold, though mostly black and white. The creator recommends a pattern like this for best results: you can also specify color XXX of pants and YYY of pattern such as XXX KangKengChang pants with YYY pattern XXX KangKengChang pants with YYY elephant pattern XXX KangKengChang pants with big YYY elephant pattern Use a weight of 0.7 and a base like <chilloutmix>
clothing	thai-school-uniform		clothes asian	A traditional Thai school uniform. Check the triggers for three tie/bow options. Works best with a very low weight of 0.2 and a model like <anylora>
lora	therock		characters realistic male	No
style	thicklines		art style anime illustrations effects	Reduces realism for anime-based models and makes the colors more vibrant and anime-like. Makes outlines more thicker, very helpful for models with washed out colors such as av4 and av45. Strength can vary from 0.1 - 2, more strength = more anime-like
lora	thigh-sex			No
lora	thighgap		anatomy porn realistic	No
lora	tia-carrere		asian chinese female realistic characters	A likeness of Asian actress Althea Rae Duhinio Janairo aka Tia Carrere. Works well with realistic base models like <believe> and similar. Will default to her prime years, resembling hear appearances on MTV and Waynes World
clothing	tifa-costume		gaming clothes final fantasy	No
lora	tim-burton			No
lora	time-slider	No		No
object	tinker-toy		toys objects	Lora created by Pirate Diffusion community member EVO. Per his civitai entry: I remember FONDLY the days of my childhood where everything was fresh, the world was yet to be discovered, and something as simple as 'Tinker Toys' were able to entertain me for hours on end. This is when I thought how cool would it be to explore Tinker Toys in a completely new age with endless possibilities. This LORA, in its initial phase, may not produce a perfect replica of this childhood classic but it sure is fun to work with. This LORA is fresh off the press and I am still trying to figure out all that it can do. I will try to post more images as I obtain them. Enjoy ???? If this LORA were a steak it might be medium rare to medium. Meaning, might need to pump up the weight a bit but most models so far seem to tolerate the extra spice.
anime	titans-blackfire		characters female cartoons anime	Blackfire Teen Titans v1.0 by TheVirtualGallery # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low # Mangled limbs, general chaos ? Some LoRas can’t handle rendering with resolutions bigger than 512x512 // Also make sure your prompt isn’t too complicated for this LoRa # No visible effect ? Check if the LoRa has any triggers that need to be added in order to activate it # Good full models to try out LoRas with: anime ? <anylora> <counterfeit> <breakanime> comics ? <comics-vision> <icomix> <misstoon> semi-real ? <revani> <lyriel16> <dreamshaper6> realistic ? <majic> <bra5> <chilloutmix> <edge>
anime	titans-jinx		characters female cartoons anime	Jinx (Teen Titans 2003) v2.0 by novelProphet # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low # Mangled limbs, general chaos ? Some LoRas can’t handle rendering with resolutions bigger than 512x512 // Also make sure your prompt isn’t too complicated for this LoRa # No visible effect ? Check if the LoRa has any triggers that need to be added in order to activate it # Good full models to try out LoRas with: anime ? <anylora> <counterfeit> <breakanime> comics ? <comics-vision> <icomix> <misstoon> semi-real ? <revani> <lyriel16> <dreamshaper6> realistic ? <majic> <bra5> <chilloutmix> <edge>
anime	titans-starfire		characters female cartoons anime	Adult Starfire (Teen titans) v1.0 by Proteinique # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low # Mangled limbs, general chaos ? Some LoRas can’t handle rendering with resolutions bigger than 512x512 // Also make sure your prompt isn’t too complicated for this LoRa # No visible effect ? Check if the LoRa has any triggers that need to be added in order to activate it # Good full models to try out LoRas with: anime ? <anylora> <counterfeit> <breakanime> comics ? <comics-vision> <icomix> <misstoon> semi-real ? <revani> <lyriel16> <dreamshaper6> realistic ? <majic> <bra5> <chilloutmix> <edge>
anime	titans-terra		characters female cartoons anime	Terra - Teen Titans v1.7 by NostalgiaForever # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low # Mangled limbs, general chaos ? Some LoRas can’t handle rendering with resolutions bigger than 512x512 // Also make sure your prompt isn’t too complicated for this LoRa # No visible effect ? Check if the LoRa has any triggers that need to be added in order to activate it # Good full models to try out LoRas with: anime ? <anylora> <counterfeit> <breakanime> comics ? <comics-vision> <icomix> <misstoon> semi-real ? <revani> <lyriel16> <dreamshaper6> realistic ? <majic> <bra5> <chilloutmix> <edge>
pose	titfuck		poses boobs porn	No
pose	titty-suck		porn poses boobs	a lora that creates breast-sucking renders Recommended settings for getting good renders: # Guidance at approx 7.0+ # Weighting varies. Weighting below 1 often causes non-suck renders, but can often be quite erotic, weighting at 1.05 plus should pretty much guarantee a solid suck :-) # Works great with realistic models and non-realistic models # Works great with basic and elaborate prompts Sidenotes: # Seems to interact well with other loras # A good resolution to render at is /size:512x512 & /size:512x768 then use /more, /strength, and /size to upscale Example Prompt: /render a male teacher is kissing a woman' tits, tittysuck <titty-suck:1.01>[blurry, lowres, worst quality] <realvis20>
inversion	tom-holland		characters male realistic comics	No
lora	tomboy		female characters fetish	This lora renders images of "tomboyish" women Recommended settings for getting good renders: # Guidance at approx 7.0+ # Best weight around 1.0 and over # Works great with realistic and non-realistic models # works well with basic and elaborate prompts Sidenotes: # Seems to interact well with other loras # Render at low sizes and use /more /size /strength to upscale # Use "very short hair" in your prompt for best effect Example Prompt: /render 1girl, solo, (tomboy:1.2), very short hair, [blurry, lowres, worst quality] <photon>
lora	tomoko-shiretoko	No		No
lora	tomoko-watamote	No		No
style	toon-chroma		art style effects cartoons	This lora was trained on cartoonish images. From the creator of Mistoon. Not a lot of information was give for this lora but through testing it was found that it does have a noticeable effect, for instance I turned an anime model <meinamix> into a more traditional western animated still. Low weights below 0.5 are recommended. Try with Mistoon, meinamix and Cetus.
style	toon-style		cartoons effects art style	This lora is intended to give a western cartoonish look. It is from the creator of Mistoon so expect a similar style. In fact this lora can be used to give a similar art style to other models. Try with Cetus or Counterfeit for vivid saturated colors
style	toon3d		.base models anime cartoons art style	Samaritan 3d Cartoon v1.0 by PromptSharingSamaritan A merge of LoRAs trained on hundreds of midjourney niji expressive images with variety of emotions. It’s great for cartoonish 3d portraits and for remixing photos to give them this particular artstyle. Add 'toothy expression' for more pronounced teeth. Can only make humans and humanoids at the moment. Works really good with clipskip:2 An example prompt: /render /parser:new /clipskip:2 /size:640x832 [[monochrome, sepia, greyscale]] masterpiece, best quality, a sad old black man with gray beard and a mustache, wearing a shirt with a flowered collar and a flowered shirt, [lowres, blurry, (bad_prompt:0.8), (artist name, signature, watermark:1.4), (ugly:1.2), (worst quality, poor details:1.4)] <toon3d>
style	toonyou		.base models anime cartoons art style	ToonYou beta3 by Bradcatt A really nice-looking model that can easily produce good results with minimal prompts. It' focused on cartoon-like vibes but can do more than that. The author recommends using it without too many negative prompts since it works the best in this way. Tips: # It' good at generating images with sizes such as 640x832 or 832x640. # Try it out with clipskip:2. # To get more vibrancy with 2.5 images, I recommend treating them with this command: "/more /size:1400x1400 /strength:0.4 /guidance:15 /images:2 <toonyou> # Tags like Realistic, HDR, etc. have the side effect of forcing the model' characteristics to change.
lora	topanga		characters realistic female white	No
style	toriyama-style		illustrations anime dragonball retro art style	Triggers with toriyama akira. I suggest weight above 0.5. I've uploaded the 60 epoch and 40 epoch version. Both can work depending on the model and the desired strength. Also I've found it gives usually better results with CLIP skip 2 (except the first image, maybe that was lucky). This is trained on Anything v4.5, so use that or AbyssOrangeMix2 if you want accurate results. Other models may still work. A big difference from my other LoRA' is that this one basically requires highres fix at least at 1.5-2x size. Keep in mind this is LoRA, so you need to use it as such. It' not a standalone model and it' not a TI.
pose	towel-breasts		female poses	This lora creates images of women with a towel covering their breasts Recommended settings for getting good renders: # Guidance at approx 7.0+ # Best weight around 1.0 # Works well non-realistic and realistic models # works well with basic and elaborate prompts Sidenotes: # Seems to interact well with other loras # Render at low sizes and use /more /size /strength to upscale. # this lora is EXTREMELY horny. If you want to keep it SFW use negative prompts like "nude, nipples, nsfw" Example Prompt: /render <towel-breasts> 1girl, towel over breasts, panties, [see-through, (nipples:1.5), (nude:1.5), (nsfw:1.9), (worst quality, low quality:1.4)] <epic-realism3>
anime	toy-doll		toys anime	No
object	toy-figmas		toys objects	A LoRA inspired by the Figma figure series of highly poseable collectible toys, by model creator STMA. A base of <aom3> or similar with a weight of 0.6 is recommended, though as high as 1.15 can also work well. Recommended prompts: Positives: realistic, 3d, 3dcg, ((octane render)), fisheye Negatives: [[simple background, grey background]]
object	toy-nendroid		objects toys	No
object	toy-paseersdolls		toys objects	No
object	toy-realisticfigure		toys objects	No
lora	tracer-overwatch		characters female gaming	No
clothing	track-suit		clothes	No
effect	TradingCards		art style graphic design comics	90s Trading Card Style Trained on the art of 1995 Ultra X-Men Trading Cards. This lora can be used to consistently center your character in a 10(x) x 13(y) or 640x832 picture scale. Use <lora:TradingCards:0.1> without trigger word to help position your character in the card while minimizing the comics style itself. The Trigger words are not required. If used, they will increase the Comics style effect in your render. Can be used to represent landscapes too. Use [[watermark,text]] to remove possible magazine artefacts
lora	trump		characters realistic male white	No
lora	tsubasa-honda		characters female japanese asian	No
style	tuna-style		cartoons effects art style	No
lora	twice-mina		characters female korean asian	No
lora	twice-momo		characters female asian korean	No
lora	twice-sana		characters female realistic asian korean	No
lora	twice-tzuyu		characters female asian realistic korean	No
lora	twilek	No		No
pose	two-people		poses male	No
anime	ugly-bastard		memes anime male characters	No
style	uhd		male penis porn realistic lgbtq .base models	Unstable Homoerotic Diffusion (UHD) v1.0 by H4RV3YD3NT / gaydiffusion A great model for anyone interested in men4men renders. The author recommends using it with the following prompts: Positive: masterpiece, best quality, high quality, realistic Negative: worst quality, low quality, anime, digital illustration, 3d rendering, comic panel, scanlation, multiple views, artist name, signature, error, text, cropped
inversion	uncannime	No		No
inversion	uncannime		anime effects anatomy memes art styles	Uncannime by Kinomi is a LoRA that exaggerates waifu features, making the eyes extra big. When paired with anime models, it creates a unique art style. When paired with semi-real or realistic base models, you get creepy valley doll-like women. A fun one to play with!
effect	underwater-naked		porn effects	No
pose	underwater2		female poses environments	a Lora that produces photos of women underwater Recommended settings for getting good renders: # Guidance at approx 7.0 # Weight is best between 0.8 and 1 # Works great with most realistic and un-realistic models # Works great with basic and elaborate prompts (seems best with <edge> and <level4>) Sidenotes: # Seems to interact well with other loras including character loras # A good resolution to render at is /size:640x640 & /size:640x832 # Less weights tend to give renders with more "partially submerged" and "by the pool" shots. Fun to play around with. Basic Prompt: /render <underwater2> underwater, 1girl, <edge
lora	undressing-another			No
lora	undressing-bra	No		No
sci-fi	universe-moonspace		backgrounds sci-fi environments	Universe-MoonSpace by Cecily_cc Trigger word: Space, moon, planet Works great for generating planets from various views. Great for anything requiring a space, galaxy, planet, moon, or Nebula backgrounds.
inversion	unreal-neg		.negatives	Unrealistic Dream by Lykon is a negative suitable for realistic images, and is intended to be used with Bad Dream and other negatives like Fast negative. These three are a common combo. Example Use: [[<unreal-neg:-1><bad-dream:-1><fast-negative:-1>]]
inversion	unspeakable-horrors		.negatives	Unspeakable Horrors 4 Vectors by PirateKitty is the mildest of the powerful negative boosters. It makes subtle changes to improve your image quality (and has nothing to do with horror topics) Usage: [[<unspeakable-horrors:-1.3>]] The default weight is 0.7. The max is 2 Use it with your favorite LoRAs and Base Models
clothing	urbansamurai		objects clothes	No
style	urpm		realistic .base models semi-real porn	No
style	urpm13		realistic .base models semi-real porn	Uber Realistic Porn Merge 1.3 is Stable Diffusion' most popular porn model, with poses and capabilities only really matched by <degen> and <hassan15>. This 1.3 version is a little more like Protogen, in that it can do semi-realistic paint and anime as well as realistic adult photos. Give it a good tug! We support both the old one <urpm> and the latest. They both work differently with LoRAs, so try one or the other to see different results. There is also a LoRA extract called <urpml> that you can mix with different base models to add that ultra-porn influence anywhere
inpainting	urpm13-inpainting	No		No
style	urpmcun		realistic .base models porn oral	Uber Realistic Porn Merge: Cunnilingus is a specialized model for for that special one thing. Despite its focused purpose, many find it hard to use, so here' an example prompt to get you started: /render /wide woman licking another woman' pussy, (((cunnilingus))), ((detailed genitals)), ((detailed faces)), 2girls <urpmcun>
lora	urpml		porn	URPM Lora is an extract of Ultimate Realistic Porn Merge as a mini-model that can be mixed with any base model to amplify its nudity and sex pose capabilities. When the porn just isn't happening, try this porn booster. Note: May glitch with other LoRAs and base models that force costumes, so mess with the weights accordingly. Try it with <realvis13>
lora	uzaki-hana	No		No
clothing	v-dress		clothes	Very easy to use. You can type "red dress" or almost any color for the desired variation. It may need some tinkering with weight to work correctly (0.3-0.9). Usually, it' not necessary. It also works with anime.
lora	vaginal-depth	No		No
lora	valkyrae		characters realistic female	No
lora	vanessa		characters realistic female retro white	No
lora	vanessahudgens		characters realistic female asian	A Stable Diffusion LoRA profile based on fan art for Vanessa Hudgens, realistic version. Try it around 0.3-0.7 weight. Try it with /sampler:dpm2m /guidance:7 /steps:50 <protonova>
style	vangoghfaker		semi-real art style paintings	Van Gogh Fake Style is a painting LoRA intended for portraits of anime waifus, mashing up the classic brush strokes with Japanese modern art. It generates a great texture at about 0.8 weight, and pairs with many base models when the guidance is set to around 6
effect	vaporwave		effects art style retro environments	Vaporwave is a throwback neon retro aesthetic, inspired by the sometimes awkward style of 80-90s computer graphics. It likes to generate images of statues which can be slightly NSFW. It also likes plants and of course pools, but should be capable of adding a vaporwave aesthetic or colorset to most subjects even if it could always be better. Set your /guidance around 7 and use the Lora weight from 3.5 to 5.5, for starters There' also some creepy Backrooms found footage style pool pictures in the training set as well, so if you're into that sort of thing you might find some use out of this LoRa for that kind of imagery as well. There are over 11 triggers to choose from to change up the effect
lora	vault-interior		gaming environments interiors post-apocalyptic	No
clothing	vault-suit		clothes sci-fi	Works better with lower CFG vaules (3-6). Recommended weight range is around 0.35-0.65 depending on the model and CFG vaule. Weight could be reduced when using SD upscale or certain img2img actions. Model should know 'with pipboy3000' ; 'without pipboy3000' and regular 'pipboy3000' prompt, some extras such as 'leather armored' ; 'laserrifle' and 'gauss rifle' to some degree. extra propts are currently quite buggy and might not work properly, will require further improvements. Example promts; full body shot of a man wearing vaultsuit in a forest / a man with a beard wearing vaultsuit without pipboy3000, standing next to a bridge / a woman wearing vaultsuit with pipboy3000 in a ruined city / back shot of a woman wearing vaultsuit without pipboy3000 in a post apocalyptic desert. Extra Info: Dataset for the model was around 64 pictures including in game screenshots from my heavily modded FO4 and some pictures from the web and manually captioning images. Released as an alpha, i will try to improve it further on my free time. Zovya' RPG v3 was used for the pictures.
style	vector		semi-real .base models gaming art style	VectorArtz Diffusion is an older checkpoint base model for isometric and vector art. It was the best model of its time for making logos and icons for apps. Try with towns, landscapes, objects, and 2d characters for gaming
lora	vector-minimal		illustrations graphic design art styles	No
lora	velma-dinkley		cartoons characters	a lora that creates renders of Velma from Scooby-Doo Recommended settings for getting good renders: # Guidance at approx 7.0+ # Best weight right around 0.3 - 0.6 # Works best with non-realistic models eg. <atoz>, <toonyou> # Works great with basic and elaborate prompt # Use /recipe:easynegative when possible Sidenotes: # Seems to interact well with other loras # Render at low sizes and use /more /size /strength to upscale # There are some great opportunities here to remix your final renders with a realistic model like <level4> to get some great results! Example Prompt: /render /recipe:easynegative 1girl, solo, cartoon style, velmadinkley, <velma-dinkley:0.5>, glasses, orange sweater, red skirt [blurry, lowres, worst quality]<toonyou>
anime	velma-dinkley-sexy		characters female realistic semi-real anime	Velma Dinkley Sexy is a prototype early release from dj226. It' Velma from Scooby Doo, but realistic and pulling off her orange clothes. In our testing the anime ones came out MUCH better than the realistic ones. Use /highdef either way for best results /render /size:768x768 /clipskip:2 Velma, realistic, natural lighting, post-processing, ultra realistic, highly detailed, very detailed, cinematic, hyper realistic, intricate scenery, sharp, photo by Annie Leibovitz, film, studio lighting, detailed skin, ultra realistic, bokeh, sharp features <lora:velma-dinkley-sexy:0.3><breakanime>
inversion	verybad-negative	No		No
effect	vhs		retro effects	This lora is supposed to emulate the feeling and effects of a VHS tape, visual artifacts and all. During testing, this lora appeared to give the effect of a VHS tape the strongest. Try combining this with other era related loras. Per the author: This model is designed to give the style of an old, poor-quality VHS tape. The safetensors file should now be possible to use natively in Automatic1111! It should be used with the Stable Diffusion 1.5 base model, and I recommend increasing the strength if you want a stronger effect. Around 1.5 seems to work pretty well.
effect	vhs2		retro effects	This lora is supposed to emulate the feeling and effects of a VHS tape, visual artifacts and all. Per the author: Japanese VHS footage with an emphasis on weather forecasts. Trained on the album art of ????. It was found that this lora doesn't produce the visual artifacts but rather, blurs and desaturates the image slightly as if the picture was from a 1990s tv broadcast.
effect	vhs3		retro effects	This lora is supposed to emulate the feeling and effects of a VHS tape, visual artifacts and all. It was found that cranking the weight to 1 helped bring out the effect more, however too high weight on this lora appears to bring about exceeded memory warnings.
pose	vibrator		poses objects porn	Women using hitachi magic wand-type vibrators. The suggested weight is 0.3 - 0.4, all though author recommends higher. Mostly used with anime type base models.
lora	victoria-justice		characters female white	No
lora	vince-vaughn		male characters white	No
clothing	virgindestroyer		objects clothes memes hentai anime semi-real female	The Virgin Destroyer is a line of skimpy sweaters, failing as functional garments to keep someone warm but excelling in making them hot in the pants. Add 'garter straps' to your prompt for the leggings attachment. Try it with <kotos> with a default weight of 1, and lower the other character weights to avoid glitching with their default costumes
styles	virile1		porn male penis trans futa .base models	No
style	virile2	No	male realistic lgbtq trans .base models	Virile Reality 2 by Scratchproof if a base model, checkpoint merge of UHD, A-Z, Better Cocks, Instant Polaroid, Realistic Vision, Juggernaut, Penis LoRA, and Tgirls. Needless to say, it is a realistic NSFW checkpoint for rendering men. The creator recommends 512x768 at /guidance:6
style	vivy		.base models semi-real anime fantasy	VivCharMix v5 New Era by Calorstate <vivy> is a semi-realistic model that doesn't require many prompts to create great renders (especially if boobs are included). It mixes well with various character loras, and can turn even some anime ones into realistic-looking characters. The author states that it makes a perfect base for further remixes with a fully realistic model. Recommend negative prompts: # [low quality, worst quality, lowres, grayscale, normal quality, username, outline, poorly drawn, copyright name, character doll, scribble , (silent comic:1.4)] OR # [low quality, worst quality, normal quality, harsh shadows, grain, EasyNegative, gif artifacts, abstract background, simple, jpeg artifacts, shaded face, barren, boredom, ashen color, fill light, (border:1.4)] OR # [low quality, worst quality, normal quality, harsh shadows, grain, EasyNegative, simple, jpeg artifacts, shaded face, barren, ashen color, fill light, (border:1.4)]
lora	volleyball-uniform	No		No
style	voxels		gaming .base models art style	VoxelArt Diffusion by Fictiverse is trained on the original Stable Diffusion 1.5 papercut model. You can do 3D and isometric gaming graphics similar to CrossyRoad and Minecraft, but it is an older model, so the voxels tend to look kinda melted together a bit
lora	waffle-style		effects	No
lora	wafflestyle	No		No
style	waifu		anime .base models illustrations	Waifu Diffusion is a long running base model trained on crisp-looking anime. Compared to your other anime base model options, this one is fairly well-rounded, which unfortunately has been detrimental in making it a clear winner in any category. But it does make great waifus
pose	waifu-catcher		anime semi-real interiors poses	Puts your waifu into a fun arcade game! works best on weights 0.5-0.8. Tokens to include: crane, game, waifu catcher.
lora	walterwhite		characters male realistic white	Walter White is a LoRA inspired by Bryan Lee Cranston' character in the television drama Breaking Bad, where he plays a mild-mannered science teacher turned meth dealer. Use the trigger options to take off his glasses, make him bald, and give him two disguises. It appears to be trained on semi-real paint styles, though some images look quite real. Try it with <dreamshaper> as a base and a weight of 0.8 for starters
style	wavy		semi-real .base models art style	Wa-vy Diffusion is an early dreambooth checkpoint base model by WavyMulder, which works wonderfully with painting LoRAs. You can achieve a variety of illustration and paint styles, particularly excelling at portraits and uninhabited landscapes. Does quite well with celebrities as well. Lowering the guidance gives you looser brush strokes
clothing	wedding-dress		clothes	No
lora	weight-slider			Weight Slider by Ostris reduces or adds fat to the character. Use negative weights to make the character skinnier, or positive to fatten up. If you go too far it just looks like an explosion at a bakery. Use the base model <realvis50> or similar
clothing	weird-future-fashion		clothes sci-fi	Weird Future Fashion by MachineCode puts characters in bizarre stereotypical future fashion trends that haven't quite caught on yet, like going outside in tiny shorts and a transparent raincoat, casual spiderman suits, and elaborate dresses with fabric that shoots straight up. Try it with a base like <deliberate2>
effect	wet-chocolate	No	effects	No
clothing	wetshirt		objects clothes porn	Wetshirt is a very flexible clothing Lora that puts a wetshirt (slightly transparent, hard nips) on a character. Works very well with almost all model types (realistic and anime).
style	whimsy		cartoons effects art style	No
lora	whip-marks		bondage	Whipmarks v1.0 by OettankoPaizuri # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low # Mangled limbs, general chaos ? Some LoRas can’t handle rendering with resolutions bigger than 512x512 // Also make sure your prompt isn’t too complicated for this LoRa # No visible effect ? Check if the LoRa has any triggers that need to be added in order to activate it # Good full models to try out LoRas with: anime ? <anylora> <counterfeit> <breakanime> comics ? <comics-vision> <icomix> <misstoon> semi-real ? <revani> <lyriel16> <dreamshaper6> realistic ? <majic> <bra5> <chilloutmix> <edge>
sci-fi	witch-punk		effects sci-fi fantasy	WitchPunkAI by Konyconi is a dark but playful visual effects style that adds metallic witch hats, knobs, and glowing auras to your art, and it works on anything from vehicles to objects to people. Search our system for the other punk variation Loras. It works best with fantasy, semi-real base models like <revani> or <realisian>
style	wlopish		art style semi-real fantasy	Wlop is an art style Lora that emulates the fantasy digital paint style of Chinese artist Wang Ling. Use a base like <aohard> or <av45> with a weight between 0.6-1
lora	wonder-woman		characters female	This lora renders images of Wonder Woman (based on film version) Recommended settings for getting good renders: # Guidance at approx 7.0+ # Best weight around 0.3 # Works great with realistic and non-realistic models # works well with basic and elaborate prompts Sidenotes: # Seems to interact well with other loras # Render at low sizes and use /more /size /strength to upscale Example Prompt: /render 1girl, solo, wonder woman <wonder-woman:0.3>, [worst quality, lowres, blurry] <photon>
lora	wonder-woman-character	No		No
lora	wondergirl-young-justice	No		No
lora	wonyoung		characters korean asian	This lora is a specialized tool for creating realistic and artistic portraits of Jang Won-young, the K-pop star from Ive. The lora can generate high-quality images of Jang Won-young in various poses, backgrounds, and lighting conditions, with the option to customize the output according to the user' preferences. The lora is ideal for fans, artists, and designers who want to explore the visual possibilities of Jang Won-young' appearance and style.
style	wool12		semi-real .base models art style	No
sci-fi	wowifier		detailers sci-fi armor	wowifier v.3 by mooncryptowow Just wow. Works great. Looks great. Use some (or all) of the extra triggers for varied outcomes. The lora is oriented towards sci-fi and mechanical parts, but can be used for other things as well with lower weights.
style	wuxia		anime fantasy illustrations creatures art style	?? Wuxia V2 by chocolae Wuxia is a Martial arts style lora, with fantasy and dragons included, simple and neat composition. Let your martial arts emerge from your hands.
effect	x-ray-hentai		hentai anime effects penis	No
lora	xennos-style	No		No
lora	xev-bellringer	No		No
effect	xl-hdr		lighting detailers effects	General XL style by HK4Life is a LoRA trained on photos from Stability' XL checkpoint, photos that boost the HDR effect and add more detail to the background, bringing the aesthetic closer to that hazy borderline realistic magical look of Midjourney. Lowering the strength reduces the HDR effect but also simplifies the background
effect	xray-glasses		effects anime hentai objects	The dream every man once had: x-ray glasses. A bit unstable LoRa that needs some tweaking before making it work, but the effect isn’t bad. How to use: # Don’t render in sizes bigger than 512x512. Anything bigger than that makes the glasses fall apart. # Works best with anime base models. It was trained on <av45>, but I had the best results with <cetus>. # For the upper body x-ray effect, use a combination of those prompts: “eyewear view, glasses, pov, upper body, x-ray glasses, x-ray vision, nipples.” # Change “upper body” and "nipples" in the prompts mentioned above for either “panty” or “public hair” for the lower body x-ray shot. This one is harder to achieve.
anime	yaemiko		characters semi-real anime female genshin impact asian	Realistic Yae Miko from Genshin Impact. Recommended weights for txt2img - 0.45-0.75, higher weights may cause bad face generate
lora	yan-cutie		characters female chinese asian realistic	Yan Cutie is a fictional Chinese character frequently seen in Asian photorealism tutorials. Use with a weight of 0.7 and lower on models like <majic5>
effect	yearbook		realistic effects retro realistic	This lora is designed to emulate the style of a vintage yearbook photo. Try specific decades. Used for portraits. Try famous people or characters and see what they looked like in High School! In testing weight 0.5 worked well.
style	yiffy		semi-real .base models illustrations furry	Of the two anthro-furry character models named Yiffy that are currently in the system, this one is the semi painting-based one, whereas YiffyMix2 is more 2d anime styled. This one is trained on models such as e18 and zeipher F111, fluffyrock, and uses the bright ko-fi vae. Use it as a base model with LoRAs or on its own. Try it with /sampler:dpm2m /steps:35 /guidance:10 to nail that paint look
style	yiffymix2		anime .base models furry	Lawla' Yiffymix is a furry model capable of generating sfw/nsfw furry artworks in general with consistent quality, while having decent details in hands, clothing, background, etc. Compared with the old Lawlasmix, it' overall more anime style oriented, and doesn't need artist tags all the time (It can still change the styles if you use them anyways). It can be tricky to write prompts for the model if you're new to it, so make sure you read the tips carefully! ?Try these settings: /guidance: 6-8 /steps: 40-150 /facelift or do /more /size:896x896 to upscale Sampler: dpm2m or k_euler_a Prompting The model can make amazing results with very simple prompts so long as you use it right. It uses both danbooru and e621 tags, like 1boy, 1girl, solo_focus. It' recommended to start the positive prompt with "(furry art, uploaded on e621:1.4)" and negative prompt with "(worst quality, low quality)". Since the merge has significant amont of anime models in it, you need to give more weight to furry related tags. Use tags like (anthro furry) in the positive prompts and negatives like [human] in order to get it to make furry art. The model performs rather well when you specify the color of the fur. For instance, ((detailed red scales)), ((detailed blue fluffy fur)). Because it is <aom3> based, the M-shaped hair is common, so try prompting against it to change things up
style	yohan		anime hentai	Released back in January 2023, Yohan Diffusion is an Anime/Hentai full base model mish-mash of Anything V3 and images from the artist known as FreeStyle. It' been described as a 2.5D anime model for its semi-realistic shading, and notorious for its hentai pinup properties. It has spawned a ton of other anime merges but some of our users still prefer the OG, so we brought it back
style	yomyom-style		art styles anime	No
anime	yonkagor		characters male furry anime	YonKaGor (from "You will be gone") v1.0 by xlevovix795 # Poor quality or broken renders ? Lower LoRa weight, try those: <lora:0.8> <lora:0.5> <lora:0.2> <lora:0.01> ? Some only work with a weight this low # Mangled limbs, general chaos ? Some LoRas can’t handle rendering with resolutions bigger than 512x512 // Also make sure your prompt isn’t too complicated for this LoRa # No visible effect ? Check if the LoRa has any triggers that need to be added in order to activate it # Good full models to try out LoRas with: anime ? <anylora> <counterfeit> <breakanime> comics ? <comics-vision> <icomix> <misstoon> semi-real ? <revani> <lyriel16> <dreamshaper6> realistic ? <majic> <bra5> <chilloutmix> <edge>
lora	yor			Yor Briar / Yor Forger from Spy Family by Lykon is one of the best anime waifus. This model contains the various costumes of docile housewife or sexy assassin. Check the info link below for example prompts, as this one is not easy to prompt for
lora	yue-avatar	No		No
lora	yurisa-chan		korean asian characters female anatomy	One of the first Loras that was deleted by Civitai and reuploaded by fans, this is an illustration of a young Korean woman. Use a guidance of 7, a lora weight of 0.5 and <anylora> or <chilloutmix>
lora	yvonne		realistic characters female white	No
style	zarch		semi-real environments illustrations art style architecture interiors graphic design	Architectural illustrations of clean, dare I say brutalist urban landscapes in a light to dull pastel color palette, an experimental model by the artist calling themselves Zarch. Use the different triggers to manipulate the pallete. The first versions of the Zarch Illustration model are limited to wide angle paintings of buildings, while the followup versions (V3) have triggers for elevation, plants, streets, crowds, and deadpan illustrations of families going about their daily lives in an almost Mondrian sense. The model creator goes by Laizhende
style	zarch2		environments illustrations art style architecture interiors graphic design	Architectural illustrations, a follow up to the basic Zarch model that only offered buildings. This update has elevations, plants, streets, crowds, and deadpan illustrations of families going about their daily lives in an almost Mondrian sense. The model creator goes by Laizhende
clothing	zatanna-young-justice		illustrations comics cartoons anime clothes	LoRA for Zatanna from Young Justice by author novelProphet on Civitai. The Lora is trained on AnythingV5. Recommended weight between 0.6-0.8. Can go higher or lower depending on additional LoRAs
lora	zelda-botw		nintendo gaming zelda	No
lora	zelda-link		zelda nintendo gaming characters male	No
lora	zelda-malon		zelda nintendo gaming characters female	No
lora	zendaya		characters female black	No
inversion	zendayac		female black characters	This textual inversion creates the likeness of film/TV star Zendaya C Recommended settings for getting good renders: # Guidance at approx 7.0+ # Best weight around 1.0 # Works well non-realistic and realistic models # works well with basic and elaborate prompts Sidenotes: # Works well with other loras # Render at low sizes and use /more /size /strength to upscale Example Prompt: /render 1girl, solo, zend4y4, <zendayac> [blurry, lowres, worst quality]<photon>
lora	zinnia			Zinnia is a fictional character created by ShubbleLover. Try <zinnia> with <juggernaut>
lora	zoe-kazan		characters female illustrations	No
lora	zombies		horror creatures	Girls and zombies. This LoRa tries to find the balance between one and the other without turning the girl into a zombie. You can modulate the effect with the trigger words, the main one is 'zombie'. Recommended weight is 0.7-0.9, but should adjust it accordingly. Try it with guidance 6-9.
anime	zonezero-nicole		female characters anime	Nicole Demara (Zenless Zone Zero) by Goofy Ai Recommended settings for getting good renders: # Weight ~0.5-0.7 # Try it out with base models such as <anylora>, <cetus> or <counterfeit> # /parser:new /clipskip:2 (optional) Sidenotes: # Rather easy to use and can handle rendering in sizes such as 832x1024
lora	zoom-slider		effects	No

Pirate Difusion Lora Category Assigner

fantasy

anime

effect

sci-fi

sci-fi

style

effect

Animal

anime

anime

pose

anime

clothing

effect

anime

fantasy

character

clothing

pose

anime

anime

anime

anime

inversion

base

lora

celebrity

object

pose

effect

lora

lora

celebrity

pose

pose

pose

pose

style

style

celebrity

effect

fantasy

celebrity

background

background

anime

effect

effect

anime

anime

effect

anime

object

style

effect

style

effect

style

style

style

style

effect

pose

pose

anime

celebrity

style

inpainting

style

celebrity

style

inpainting

base

celebrity

anime

character

anime

style

celebrity